American Wood Council Welcome to STD 104: ASD and LRFD with

Transcription

American Wood Council
Welcome to STD 104: ASD and LRFD with the 2005 NDS.
Copyright © 2005-2011 American Wood Council
All rights reserved.
1
The National Design Specification® (NDS) for Wood Construction has been
based on allowable stress design since the first 1944 edition.
2
The NDS remained an allowable stress design methodology through the
2001 Edition.
3
Load and Resistance Factor (LRFD) for wood was first introduced in 1996 in
the LRFD Manual for Engineered Wood Construction. Its basis was the
AF&PA/ASCE 16-95 Standard for Load and Resistance Factor Design for
Engineered Wood Construction.
4
The industry incorporated ASD and LRFD in a single dual-format standard
with the release of the 2005 NDS.
5
In this eCourse, see how the 2005 NDS is organized and see what each
chapter contains. Also receive an overview of the LRFD concept. Along the
way, see what has changed from previous editions of the document. Worked
examples will be used to show how this document works for both the ASD
and LRFD process.
6
Publication of the 2005 Edition of the National Design Specification® (NDS®)
for Wood Construction culminated 3 years of development by AF&PA’s ANSI
standards development committee dedicated to providing state-of-the art
information for wood design. The 2005 NDS was approved as an American
National Standard on January 6, 2005 with a designation ANSI/AF&PA NDS2005.
There are a few changes and additions for the 2005 document.
7
In the 2005 NDS, chapters have maintained the same order as the 2001
NDS to provide a more comprehensive document for the design of wood
products for building construction. Chapters are grouped in a logical fashion
beginning with general provisions, then wood materials, connections, and
finally special assemblies and provisions.
8
2005 NDS Appendix E maintains its importance for checking local stresses
in fastener groups. Appendix N is the only mandatory appendix and is new. It
contains all of the necessary tables to apply LRFD in the NDS.
9
The NDS Supplement: Design Values for Wood Construction, an integral
part of the NDS, has also been updated to provide the latest design values
for lumber and glued laminated timber.
10
Now, let’s review the LRFD concept.
11
So what is load and resistance factor design? We’ll discuss the design
process, the design concepts, and a comparison with allowable stress
design.
12
The underlying basic philosophy for the process of structural design is that
the capacity of the structural system must exceed whatever demand is
expected on the system.
13
The structural design process fundamentally breaks down into five key
components as shown here. Others issues including fire, economics, and
aesthetics are handled separately. The demand includes type, magnitude,
and placement of loads on the system and the resulting interaction with the
system’s geometry. The capacity of the system is provided in combination by
judicious choice of materials, section geometry, and an understanding of the
way the system behaves under load. The subject matter of this seminar will
be dealing with the capacity side of the process - featuring wood as the
material.
14
A limit state is the point at which the structure fails to serve its intended
purpose in some way. Two broad limit states can be identified for structures:
safety and serviceability.
15
Serviceability limit states appraise the structure in terms of its everyday
usefulness. For this reason, it is important to know how well the structure is
actually performing. A way of seeing this, is to consider average material
strength values in combination with real load magnitudes in the measure of
actual performance.
16
Safety on the other hand can be thought of in statistical terms - probability of
failure, or conversely, survival. Using statistics, one can appraise the safety
of a structure in terms of measurable probability. In the LRFD method, the tie
to a statistical approach is achieved through the use of load factors and
material reference strengths modified by reliability factors.
17
Here is a symbolic representation of the structural property variability among
several wood products. Plotted here is the relative frequency of occurrence
against actual property values from testing. Structural testing in specific
modes is performed on these products to produce the data set that makes
up these curves. Each curve (normal distributions shown here) can be
described by statistical measures including mean and standard deviation (a
measure of the spread of the curve).
18
For this normal distribution curve, it is assumed that the probability of
occurrence equals 100%. From this, one can determine, for example, the
structural property value that is appropriate for 5% of the sample population
(common for wood strength properties).
19
Let’s take two distributions: one for load (S), and one for resistance (R); and
plot them together. Each of the curves has its own unique statistical
description (mean and standard deviation), and may or may not have the
same distribution type. Normal distribution types are shown here, but there
are others, chosen to best fit the test sample data points. Note that the
resistance curve is to the right of the load curve, and the curves overlap. The
overlap implies the region where load is greater than or equal to resistance,
hence potential for failure.
failure
20
The overlap, or failure zone, can be represented in a more useful way. If the
load and resistance distributions respectively are normalized to the same
type, then a performance distribution Z can be created by subtracting the
load distribution from the resistance distribution. The statistics of Z are
determined as seen in the slide, as well as fZ itself. In this plot, the area
under the fZ distribution that falls in the region of property values less than
zero, represents the probability of failure of the structure in this particular
mode of testing
testing. Now a measurable probability of failure is available
available. It can
be further described in terms of the number of standard deviations away
from the mean of the performance distribution. The Greek letter , known as
the safety index, is used to describe this multiple. Thus,  is directly tied to
the probability of failure.
21
For large values of , the probability of failure is very small. For small 
values, the probability of failure is much larger.
22
These are typical  values used in structural design in various materials,
including wood. It is interesting to note the corresponding probability of
failure. These are levels for which designers have historically been designing
buildings.
23
How is  actually used in design? Beta is actually invisible in the design
process. It is tied to two other factors: the reliability index  (used on the
capacity side of the equation), and the load factor  (used on the demand
side of the equation). To design for any demand with any material to a target
, it is prudent to fix the value of the load factor  (standardized values for all
materials), and derive reliability indices  for various structural properties of
various materials. This process is known as calibrating the reliability index.
Calibration needs to cover all of the relevant factors such as the load and
variability of the member strength based on species, grade, and type of
application. Generally for wood, the 5th percentile of the strength test data is
used for the resistance side, while load statistics are obtained from extensive
studies of structures in all climatic zones and with different occupancies.
24
A calibration example: the bending strength of 2x8 lumber subjected to
Quebec City snow load. What  value would be appropriate for a target  of
2.6?
25
In this - ( fixed) correlation plot, the Quebec City snow load is modeled
with a lognormal distribution, while the bending strength of 2x8 lumber is
modeled with four different distributions that are fit as closely as possible to a
complete data set of full-sized test results. To give a target  of 2.6,  would
range from 0.55 to 1.0 depending on which mathematical model is used for
the resistance. This shows how sensitive  is to the assumed distribution
type.
26
Here is a cumulative probability plot of 2x8 bending strength. On the plot is
the complete test data set of full-sized specimens (In-Grade) and two
distribution models that are fit as closely as possible to the test data. The
test data comes from the 5th percentile modulus of rupture.
27
Careful inspection of the strength test data reveals that, while the 100%
distribution curve fits the complete data set reasonably well, the model
doesn’t represent the lower end of the data set very well. The lower tail is the
most important portion of the test population since the low strength members
are the ones most vulnerable to failure. Another distribution model can be
chosen for use in the calibration to better represent the lower end of the test
data set (the lower 15%). This will ultimately produce a much narrower range
of  values.
values
28
Re-plotting the - ( fixed) correlation using the lower tail model yields a
better result. In this case, the value of  = 0.85 used for bending strength is
consistent with that found in the design code equation. The procedure to
calibrate the code values with a probability analysis is mathematically
sophisticated, and is not typically part of the design process. It is useful
however to be aware of the background to the design rules to gain a better
understanding of issues affecting safety and reliability.
29
How different is LRFD from ASD in terms of a design process?
30
Many of the ASD features that designers have come to know remain the
same for LRFD: basic equation format, adjustment factors, behavioral
equations. In terms of application of LRFD principles, design process does
not change much. The demand side requires unfactored and factored (new)
load calculations. The capacity side remains in the same form. Procedural
steps are essentially the same as ASD for various structural components.
31
What does change between ASD and LRFD?
32
These are some of the distinguishing features of LRFD. There is new
notation. Calculations will develop bigger numbers as end-results. And there
is a terminology change.
33
Here is why you get bigger numbers with LRFD in design calculations. The
way safety is addressed in the two approaches is fundamentally different.
ASD makes use of a theoretical safety margin that is applied to material
stresses.
34
Factored load equations (with few exceptions) are standardized across all
material groups. Resistance values are only modified by a reliability factor
that varies by material and mode of use.
35
Here are a few factored load combinations used for safety analysis in LRFD.
Note that load factoring accounts for the probability of multiple transient live
loads occurring on the structure simultaneously. In certain cases, load
factoring leads to greater efficiencies in the design process over unfactored
ASD loads.
36
The LRFD resistance factors (or reliability indices) for wood are shown here
from NDS Appendix N for member properties and connections. The lower
the number, the more variable the material in the respective mode.
37
LRFD introduces a new terminology called the time effect factor, formerly
known as load duration (CD) in ASD. The time effect factor, , is an
adjustment for the effects of load duration and is calibrated to the primary
load in a given load combination. LRFD also employs a new baseline
(calibration point) of 10 minutes versus 10 years for ASD. Reduced to 3
general factors: 1.0 for short term, 0.8 for long term, and 0.6 for permanent;
this approach is consistent with international codes. By prescription,  is tied
to the LRFD load combination equation used
used.
38
The factor KF, included in NDS Appendix N, converts ASD material values
from the 2005 NDS Supplement for use with LRFD. This makes
implementation of LRFD within one document very straight forward.
39
Here is the format conversion factor table from NDS Appendix N. It is
dependent on the wood property, its reliability index, and application. The
resistance factor, φ is used to adjust the LRFD design value for variability.
However, reference design values in the NDS are based on near-minimum
values and thus are already adjusted for variability. In order to be used with
conventional LRFD design procedures, the NDS reference design values
(ASD based), must be divided by the resistance factor. The NDS Format
Conversion Factor
Factor, KF , explicitly divides a constant conversion factor by the
resistance factor consistent with the format used in ASTM D5457 Standard
Specification for Computing Reference Resistance of Wood-Based Materials
and Structural Connections for Load and Resistance Factor Design.
40
The question of why use LRFD for wood often arises. There are advantages
for LRFD including designing in multiple materials which may have an LRFD
basis. There is a more rational treatment of loads with LRFD, and
efficiencies often result because of this. Further, ASD load combinations
have not been maintained in ASCE 7 in deference to LRFD load
combinations.
41
Here’s an example comparing the two design processes. Consider a simple
beam under uniform load, with given section properties. We have a
displacement limit state (maximum) of span/360. Both methods require
determination of safety and serviceability loads. Note the inclusion of
prescribed load factor(s) in the LRFD load. The serviceability loads are the
same for both approaches.
42
Here we consider two safety limit states: shear and flexure. The demand /
capacity relations for shear for this problem are shown. ASD modifies the
capacity with the CD factor for load duration. The LRFD capacity equation
includes the time effect factor, , and the reliability factor for shear, v, as
well as the KF format conversion factor. Note that factored LRFD loads are
used.
43
The demand / capacity relations for flexure for this problem reveal much the
same in comparison. ASD modifies the capacity with the CD factor for load
duration. The LRFD capacity equation includes the time effect factor, , and
the reliability factor for bending, b, as well as the KF format conversion
factor. Note again, that factored LRFD loads are used.
44
The serviceability limit state considered here is maximum displacement of
span/360 under service load, wL. Note that both approaches use identical
equations. The important note here is that LRFD uses unfactored loads just
like ASD.
In summary, the design process for wood has not changed. LRFD requires
use of load and resistance factors with which designers presently skilled in
steel and concrete design using LRFD already are familiar. But as will be
seen, there are advantages to be gained with LRFD in final section
determination, especially if the problem is governed by a safety limit state.
45
Next, we’ll see what each chapter contains.
46
Chapter 1 describes terminology used in the NDS. For 2005, because of the
dual ASD and LRFD format, there are two changes. The old term “allowable”
has given way to adjusted. And the base design values from the NDS
Supplement are now called reference design values. Reference values are
those without adjustment factors applied. Adjusted values are reference
values with adjustment factors applied.
47
Chapter 1 contains reference information from relevant load documents. This
version of the NDS references ASCE 7 – 02.
48
Chapter 2 deals with adjustment factors that are global in origin. These
factors often are representative of the environment in which the wood
structure is placed. Wet service, temperature, and load duration may
become critical issues depending on the environment. The new feature here
is the time effect factor for LRFD notated as  found in Appendix N.
49
Let’s take a look at the wet service factor.
50
Design values tabulated in the NDS for sawn lumber apply to material
surfaced in any condition and used in dry conditions of service. Such
conditions are those in which the moisture content in use will not exceed a
maximum of 19%. The graph, here, shows how wood in the right conditions
of environmental temperature and relative humidity can reach equilibrium
moisture content (EMC’s) of 19% or more. This >19% regime not only
requires adjustment of some of wood’s structural properties, but can also
create an environment for durability issues
issues.
51
Here graphically, in somewhat general terms, is what happens to various
structural properties of wood in the region of high EMC. Decreases in
structural properties are noted, especially for the crushing strength.
52
Wet service adjustment factors are provided for uses where the 19% EMC
limit will be exceeded for a sustained period of time, or for repeated periods.
Applications in which structural members are regularly exposed directly to
rain and other sources of moisture are typically considered wet conditions of
service. Members that are protected from the weather by roofs or other
means but are occasionally subjected to wind blown moisture are generally
considered dry applications. The designer must use discretion.
53
Chapter 3 describes the behavioral relations used in designing wood
structures. Additionally, it describes the process of obtaining adjusted
stresses from reference values. Between the ASD and LRFD processes,
there is not much difference. There are a few additional factors for LRFD to
deal with format conversion and safety issues. Adjustment factors are used
to deal with wood in specific applications, for which values are found in the
respective material chapters that follow.
54
One of the changes in this chapter for 2005, is the restructuring of CL for
beam stability. The critical buckling term, FbE, is rewritten as a function of
Emin, the fifth-percentile modulus of elasticity value.
55
Because the design equation for KbE included a reduction for safety, two
different formats of the 2001 NDS equation would have been needed to
address both ASD and LRFD. Instead, the 2005 NDS utilizes Emin, which is
adjusted for safety, so the safety factor is not part of the basic design
equation. Applicable adjustments to Emin, based on applicability of
adjustment factor tables are used to establish the appropriate adjusted
modulus of elasticity for beam and column stability, E’min for either ASD or
LRFD.
LRFD
56
Similarly for columns, the critical buckling term, FcE, is rewritten for
applicability to both ASD and LRFD.
57
The background justification for this change is identical to that for the beam
equation.
58
For sawn lumber and glulam, reference modulus of elasticity for beam and
column stability, Emin, which represents an approximate 5% lower exclusion
value on pure bending modulus of elasticity, divided by a 1.66 factor of
safety, can also be calculated using the above equation.
59
Here’s a numerical column example comparing ASD and LRFD design
processes. Both ASD and LRFD methods require determination of loads.
Note the inclusion of the prescribed load factor, , on the load side of the
LRFD equation.
NDS Table 5.3.1 outlines applicable adjustment factors for glulam. ASD
modifies the compression capacity with the CD factor for load duration. The
LRFD capacity equation includes the time effect factor , the reliability factor
for compression c, and the format conversion factor, KF. Other adjustment
factors will be discussed later, but are identical for ASD and LRFD.
60
Consider a pinned column under axial load, with given section properties.
This 16 foot unbraced column has applied dead and live loads.
61
First, compute loads for safety design. LRFD uses load factors applicable to
the load type and typically results in a larger numerical value than ASD.
62
For this design, try a 6¾” x 9” glulam (combination symbol 1). From section
geometry, cross-sectional area is found. The column can buckle through the
X-X or Y-Y directions depending upon bracing present in each direction. It is
important to check bracing geometry and its relationship to section
dimension. Here, the column is unbraced over its entire height, so the
column could buckle in the direction of least section dimension (in the Y-Y
plane here). Checking slenderness ratio gives an appreciation for this. In this
case the least dimension direction
case,
direction, b
b, governs
governs, with a slenderness ratio of
28.
63
Next, consider the column’s environment. Any changes from the base
environment for the wood must be reflected in the adjustment factors. Note
the adjustment factors for load duration are different numerically for LRFD
and ASD.
64
For material design values, the 2005 NDS Supplement contains reference
values for compression, E, and Emin. Note that values are the same for both
ASD and LRFD. The column design parameter “c” for glulam is 0.9. The
LRFD resistance factors for column compression and stability and the format
conversion factors KF are taken from Appendix N of the standard.
65
The first limit state for columns is crushing per NDS 3.6.3. Compute the
crushing strength using Fc* which includes all adjustment factors except the
column stability factor, CP. The allowable crushing load is computed as
shown.
66
The second limit state is buckling. The column buckling equation is derived
from the familiar Euler formulation simplified further here for rectangular
sections. Note that the buckling stress is higher for LRFD than for ASD.
67
The slenderness of the column lies somewhere between the crushing and
buckling limit states. To find out where, compute the column stability factor,
Cp. This factor reduces the crushing strength based on the slenderness of
the column.
68
The expression of the Cp equation is exactly the same for both ASD and
LRFD. Note that the column stability factors are comparable but not
identical. Capacity of the column is computed using the expressions shown.
Note again, the LRFD capacity value is higher.
69
Now compare the loads to the column capacities for each method. The trial
column works for this design for both methods. No doubt, LRFD has higher
numbers, but we can see the approximate equivalence in the two methods
through the load/capacity ratio. The two ratios are exactly the same for both
ASD and LRFD processes. Overall, the design process for LRFD is
remarkably similar to ASD.
70
For tension parallel-to-grain members, behavioral equations don’t change
and the format is the same as that for bending and compression members.
71
Tension perpendicular-to-grain is wood’s weakest link and should be avoided
per NDS 3.8.2. Awareness of how the wood is being loaded is needed to
avoid this issue. Notches, moment connections, or hanging loads below the
neutral axis can initiate these stresses.
72
The interaction equation for combined bi-axial bending and axial
compression for wood members incorporates three components. The ratio of
actual to adjusted compression stress is squared based on tests of short
beam-columns. The moment magnification factor is shown in the
denominator of the bending portions of the equation. These adjustments are
consistent with similar adjustments for other structural materials and are
based on theoretical analysis verified by tests.
73
As with the beam and column equations, Emin appears as a variable in all of
the Euler terms.
74
The provisions for bearing perpendicular to grain are the same as those in
earlier versions of the NDS. Research indicates that the smaller the width of
the plate or bearing area relative to the length of the test specimen, the
higher the proportional limit stresses. Therefore, a bearing area factor, Cb, is
used to increase the capacity for cases like washers, metal straps, hangers,
or studs bearing on wood sills or bottom plates.
75
Chapter 4 begins the wood material chapters. Design values for visually
graded and mechanically graded lumber, timber, and decking are referenced
in the NDS Supplement: Design Values for Wood Construction, which will be
covered in more detail later.
76
Adjustment factors are unique to the material described in the chapter. All
adjustment factors appear in basically the same format, but include factors
unique to ASD and LRFD in addition to factors applicable to either
methodology. Factors unique to lumber and not already discussed include
size, flat use, incising, repetitive member, and buckling stiffness factors.
77
Size and flat use factors are shown in the NDS Supplement for lumber,
timber, and decking with certain exceptions.
78
Adjustment factors for incising and buckling stiffness are provided in NDS
4.3 and 4.4.
79
The repetitive member adjustment factor is provided in the NDS Supplement
for dimension lumber.
80
The removal of the form factor stems from the fact that this value was
originally derived from plastic deformation in small clear specimens that may
not be applicable to full-size members. In addition, its applicability to
standard wood products (which are almost always rectangular in crosssection) was limited. The form factor is not allowed in poles & piles since it is
already built into reference design values.
81
An example for tension design values for an unincised axially loaded tension
member in a normal environment as defined in NDS 2.3 and 4.3 is shown for
both ASD and LRFD.
82
Finger-jointed lumber has gained wide acceptance in the building and construction industry.
Thi product
This
d t iis accepted
t d ffor use under
d b
both
th th
the International
I t
ti
l Building
B ildi Code
C d (IBC) and
d th
the
International Residential Code (IRC), and is considered interchangeable with solid-sawn
dimension lumber of the same size, grade, and species. The codes use the term “endjointed lumber” which is a generic term for lumber formed by gluing smaller pieces together
end-to-end. One manner of making that connection, the most common, is finger jointing.
Two new ASTM standards were developed: D 7374 Standard Practice for
Evaluating
E
l ti El
Elevated
t dT
Temperature
t
P f
Performance
off Adhesives
Adh i
U
Used
d iin E
Endd
Jointed Lumber and D 7470 Standard Practice for Evaluating Elevated
Temperature Performance of End-Jointed Lumber Studs for the evaluation of
adhesives used in end-jointed lumber. Products joined with qualified heat-resistant
adhesives include the designation HRA in the grade mark. Finger-jointed lumber joined with
other adhesives is marked as NON-HRA. Finger-jointed lumber without HRA designations in
the grade stamp are being considered as produced with Non-HRA adhesives. These
products
d t should
h ld nott b
be used
d iin assemblies
bli where
h
fifire-resistance
i t
ratings
ti
are required,
i d unless
l
additional testing has been conducted to demonstrate compliance.
Potential exposure to moisture and load conditions also impact the type of finger-jointed
product used.
83
The grade marks on a finger-jointed piece of lumber are very similar to those
on solid sawn lumber. There are, however, some differences. IBC Section
2303.1.1 Sawn Lumber, states, “Approved end-jointed lumber is permitted to
be used interchangeably with solid-sawn members of the same species and
grade.” The new HRA marks are intended to provide regulators and users
additional information to identify which finger-jointed products meet elevatedtemperature performance requirements.
HRA-marked finger-jointed lumber should be used for assemblies that
require a fire resistance rating under the IBC and IRC. Typically, fire ratings
are required for multi-story or multi-family structures in separations between
living units. Common walls in commercial structures may also require fire
rated assemblies.
84
NON-HRA grade marked lumber is generally permitted in residential
construction. Under current building codes, detached single-family homes
rarely require fire rated assemblies. NON-HRA marked lumber and fingerjointed products with no HRA designations can continue to be used in
construction where no fire rating is required.
85
Finger-jointed products with no HRA designations are treated as NON-HRA.
However, the material supplier should be given the opportunity to
substantiate the type of adhesive used in unlabelled material. It is possible
that qualified adhesive was used in the manufacture of the joints, but the
label was not applied.
86
Structural finger-jointed lumber is manufactured to meet the requirements of two different
t
types
off end-use
d
applications.
li ti
Th
The fifirstt category
t
iis b
basically
i ll an all-purpose
ll
product
d t indicated
i di t d
by CERT EXT JNTS on the grade stamp. The second category is appropriate for use where
the long-term loading will be primarily in compression, as indicated by VERTICAL USE
ONLY on the grade stamp.
Finger jointed lumber grade-stamped CERT EXT JNTS is intended for ALL structural
applications subject to any additional fire rating requirements. This lumber is assembled with
a waterproof
waterproof, exterior-type
exterior type adhesive
adhesive, meeting the requirements of ASTM Product Standard
D2559. Limitations on knot size and placement near joints is highly restrictive, and testing
and quality control procedures are also rigorous.
The exterior-type adhesives for CERT EXT JNTS products are suitable for bonding
structural end-jointed and laminated wood products for use in general construction where a
high strength, waterproof adhesive bond is required. Long lengths, up to 32’ or more, are
one
o
eo
of tthe
ed
distinct
st ct ad
advantages
a tages o
of st
structural-glued
uctu a g ued finger-jointed
ge jo ted p
products.
oducts Thiss lumber
u be may
ay
be used as beams, joists, rafters, studs, plates, or in any other exterior or interior framing
application. The species and grade indicated on the stamp can be expected to retain the
same structural properties as its solid-sawn lumber counterpart.
As an example, here’s a grade mark of a finger-jointed piece of lumber. It contains the same
information as is required for solid sawn lumber. In this case, the glue used in the joints is
suitable for exterior use and this is stated by EXT JNTS. Also note that since the HRA /
NON-HRA mark is missing, this material should be considered as NON-HRA.
87
There is a geometric condition in which there is a limitation on the use of some finger jointed lumber. Some lumber
may be labeled for VERTICAL USE ONLY or STUD USE ONLY as you see here
here. In this case
case, the glue in the
joints is of a type that may creep under long-term bending load. Studs labeled as such should be used only for
long-term vertical loading (axial compressive loads). They should not be used in applications where sustained
bending is the dominant load to be resisted.
A question comes up at times about whether this limitation prohibits these studs from use in walls that are subject
to high-wind or seismic loads since those walls see lateral loads which could induce bending in the studs. These
bending loads are always of short duration, well within ability of these finger-jointed studs to resist them.
These products are typically assembled with a water-resistive adhesive (indicated on the grade stamp as CERT
GLUED JNTS). VERTICAL USE ONLY products indicated as CERT GLUED JNTS are limited to conditions where
the glued joint will not be exposed to repeated wetting and the moisture content of the wood will not exceed 19%
in use. This is the most common form of this product, and VERTICAL USE ONLY finger jointed lumber lengths are
limited to 12 feet. However, customers may occasionally find VERTICAL USE ONLY products indicated as CERT
EXT JNTS in the marketplace when a mill manufactures VERTICAL USE ONLY under a recognized CERT EXT
JNTS program using waterproof exterior-type adhesives.
Note that studs with any of these grade marks are also considered NON-HRA, even though it’s not a requirement.
As a handling advisory, although structural finger-jointed lumber grade stamped VERTICAL USE ONLY – CERT
GLUED JNTS is assembled with water-resistant adhesives, these products should not be stored where water
might collect in a stack of lumber for an extended period. If the material does get wet during storage or delivery, it
should be separated so it will dry, or be installed so it may dry in place.
For more information on finger-jointed lumber products, testing, and standards, see: Western Wood Products
Association TG-9: Structural Glued Lumber.
88
Finally, a word about grade marks as might be seen on lumber in the field.
Each piece of lumber should have a grade mark that states that it is fingerjointed. But because the finished piece of lumber is composed of smaller
pieces joined together, it’s possible that some of the smaller pieces may
have a grade mark on them that originally applied to the piece of lumber
from which they were cut. Those old grade marks are supposed to be
obliterated as seen here (rubbed out black square in top of picture) but
sometimes are missed in the process
process. Old grade stamps can be ignored
since the grademark for the finger-jointed lumber is applicable.
89
Chapter 5 for structural glued laminated timber (glulam) for 2005 adds new
design process capability, increased shear strength, and new materials.
90
The radial tension adjustment factor Frt is now included in the adjustment
factor table. Radial tension is often a design consideration in curved or
arched glulam members. The table has also been reformatted to include
adjustment factors unique to ASD and LRFD in addition to common ones.
91
The volume factor, CV, for glulam retains its 2001 NDS form. It includes
terms for the effects of width, length, and depth and is based on ASTM D
3737. It applies when glulam bending members are loaded perpendicular to
the wide face of the laminations and is not applied simultaneously with the
beam stability factor, CL. The smaller of the 2 adjustment factors applies.
92
Another factor applicable to glulam and not discussed earlier includes the
curvature factor, Cc. The reference bending design value is adjusted for the
curved portion of the bending member only – not the straight portion.
93
An example of compression parallel to grain design values for an axially
loaded compression member in a normal environment as defined in NDS 2.3
and 5.3 is shown for both ASD and LRFD.
94
Round timber poles are typically used in post-frame construction. Round
timber piles are generally used as part of a foundation system. Poles
typically have the larger (butt) end embedded in the ground while piles
generally have the smaller (tip) end driven into the ground.
95
Design values for piles are based on ASTM Standard D 2899. Design values
for poles are based on ASTM Standard D 3200. Timber poles supplied to
the utility industry are graded according to ANSI Standard O5.1, therefore if
they are to be designed per the NDS, they must be regraded in accordance
with ASTM D 3200. There are no changes to the poles and piles design
values from the 2001 NDS.
96
The adjustment factor table has been revised to incorporate LRFD
adjustment factors in addition to traditional factors applicable to poles and
piles.
97
Other factors applicable to poles and piles and not discussed earlier include
untreated, critical section, and single pile factors. These factors are outlined
in NDS Section 6.3.
98
An example of compression parallel to grain design values for a single,
axially loaded, treated pole or pile, fully laterally supported in 2 orthogonal
directions, used in a normal environment as defined in NDS 2.3 and 6.3 is
shown for both ASD and LRFD.
99
Chapter 7 applies to engineering design with pre-fabricated wood I-joists
conforming to ASTM D 5055. No changes were made to design provisions
from the 2001 NDS provisions from the I-joists chapter. Designers are
encouraged to consult proprietary design information for the product under
consideration.
100
Adjustment factors for I-joists are similar to those for other wood products.
Adjustments for LRFD methodology have been added as shown in NDS
Table 7.3.1.
101
A significant change in this chapter is the repetitive use factor returning to
unity. This was revised to agree with a change in ASTM D5055-02, and is
maintained for clarity transitioning from the 2001 NDS.
102
Examples of design values for fully laterally supported bending members
loaded in strong axis bending and used in a normal building environment
(per NDS 2.3 and 7.3) are shown for both ASD and LRFD.
103
Chapter 8 structural composite lumber (SCL) have extremely low variability.
There are no changes to this section from the 2001 NDS. Designers should
consult proprietary information for the product under consideration.
104
Adjustment factors for SCL are similar to those for other wood products.
Note that similar to glulam, the volume factor is not cumulative with the
lateral stability factor.
105
The repetitive use factor Cr remains at 1.04 for 2005. It is different in value
than lumber and is applied only to the bending stress if three or more
members are sharing load in close proximity.
106
Examples of design values for a single fully laterally supported bending
member loaded in strong axis bending and used in a normal building
environment (per NDS 2.3 and 7.3) are shown for both ASD and LRFD.
107
Design values are expressed algebraically in typical terms in this chapter for
wood structural panels, but numerical design values need to be obtained
from an approved source.
108
The adjustment factor table includes adjustment factors for product
fabrication as well as size for both ASD and LRFD processes.
109
The grade and construction adjustment factor is shown as a multiplier for
Structural I panels in the ASD/LRFD Manual for Engineered Wood
Construction Chapter M9 tables.
110
Other adjustment factors for panel size, wet service, and temperature are
shown in NDS Commentary section C9.3.
111
Examples of design values for a non-structural I wood structural panel,
greater than 24” in width, loaded in bending, and used in a normal building
environment (per NDS 2.3 and 9.3) are shown for both ASD and LRFD.
112
Chapter 10 begins the connections chapters. Design issues such as
evaluating stresses in members at connections, eccentric connections, and
mixed fasteners are discussed. Reference design values are contained in
subsequent chapters, however, design for single and multiple connectors
and application of adjustment factors are discussed.
113
For mechanical connections, the term “full design value” is revised with the
intent that mechanical fasteners should be appropriately placed so that they
can develop their full design value capability as tabulated in the 2005 NDS.
In order to do this, and assure proper placement, the provisions of NDS
Appendix E to check local stresses should be used.
114
These provisions show the changes made in the 2005 NDS from the 2001
NDS to make the terminology more clear in intent.
115
As in the 2001 NDS, provisions for stresses in members at connections have
been written as follows:
10.1.2 Structural members shall be checked for load carrying capacity at
connections in accordance with all applicable provisions of this standard
including 3.1.2, 3.1.3, and 3.4.3.3. Local stresses in connections using
multiple fasteners shall be checked in accordance with principles of
engineering mechanics. One method for determining these stresses is
provided in Appendix E.
116
The 2001 Edition of the National Design Specification (NDS) for Wood
Construction contained editorially clarified provisions for checking stresses in
members at connections. The following requirements, included in the 2005
NDS, are also applicable to all prior editions of the NDS:
Stresses in Members at Connections - Structural members shall be checked for
load carrying capacity at connections in accordance with all applicable provisions of
the NDS. Local stresses in connections using multiple fasteners shall be checked in
accordance with principles of engineering mechanics.
One method for determining these stresses is provided in Appendix E from
the 2005 NDS, which is also available free from www.awc.org. All referenced
sections and design values used in sample solutions of this Addendum are
based on information in the 2005 NDS.
117
Through testing, it was learned that where a fastener group is composed of
closely-spaced fasteners loaded parallel to grain, the capacity of the fastener
group may be limited by wood failure at the net section or tear-out around
the fasteners caused by local stresses.
118
By increasing the spacing between the fasteners, much higher capacity and
ductility is achieved, even with fewer fasteners!
119
Tabulated nominal design values for timber rivet connections in Chapter 13
account for local stress effects and do not require further modification by
procedures outlined in Appendix E. The capacity of connections with
closely-spaced, large diameter bolts has been shown to be limited by
the capacity of the wood surrounding the connection. Connections with
groups of smaller diameter fasteners, such as typical nailed connections in
wood-frame construction, may not be limited by wood capacity.
Appendix E leads the designer through the stress checks for three failure
modes: net tension capacity of the wood through the cross-section and row
tear-out are the first two modes.
120
The third mode is group tear-out.
Modification of fastener placement within a fastener group can be used to
increase row tear-out and group tear-out capacity limited by local stresses
around the fastener group. Increased spacing between fasteners in a row is
one way to increase row tear-out capacity. Increased spacing between rows
of fasteners is one way to increase group tear-out capacity.
However, Footnote 2 of Table 11.5.1D (2005 NDS) limits the spacing
between outer rows of fasteners paralleling the member on a single splice
plate to 5 inches. This requirement is imposed to limit local stresses resulting
from shrinkage of wood members. When special detailing is used to address
shrinkage,
g such as the use of slotted holes, the 5 inch limit can be adjusted.
j
These provisions apply to the 2005 NDS and ALL PRIOR EDITIONS. The
example calculations provided in Appendix E use design values from the
2005 NDS. Appendix E in its entirety is available as a free PDF download
from www.awc.org.
121
Now, let’s work a complete example for a truss bottom chord splice. The
bottom chord has a tensile force of 20,000 lbs based on dead and live loads.
Other environmental conditions are as shown.
122
Efficient choice of a trial section requires practical, as well as engineering
considerations. For example, choice of lumber species, grade and even
commonly available sizes may differ among geographic regions of the
country. In addition, other considerations include dimensional compatibility
with the other members of the truss or minimum sizes required to adequately
connect the truss members (while meeting fastener edge distance
requirements).
In this examples, the chord includes connections with two rows of 7/8 inch
bolts (in a 1/16 inch oversized hole) spaced per NDS Section 11.5 for full
design values. Check the local stresses to verify member size selection.
123
Using Selection Tables: Select a member(s) from the tension member
selection Table M4.5-1a in the ASD/LRFD Manual for Engineered Wood
Construction that is adequate to resist 20,000 lbs tensile force (T) due to
combined dead load and occupancy live load (D+L). Try 4x12 No.1 Hem Fir.
T’ = 24,600 lbs.
124
Now for the splice connection. To simplify, consider a single shear
connection using one steel splice plate and neglect eccentricity in the joint.
Set the rows at the 1/3 depths which is well within NDS spacing limitations.
125
Using NDS Appendix E provisions, calculate local stresses in the fastener
group:
Net Section Tension
The net cross sectional area is calculated as (3.5)(11.25-(2)(0.9375)) = 32.8
square inches.
ZNT’ = 625(32.8) = 20,500 lbs > 20,000 lbs OK
126
Row Tear-out Capacity
From the NDS Supplement, Fv’ = 150psi. Critical spacing is the lesser of the
end distance (7D here for full design value), or the spacing between
fasteners in a row (4D); in this case, 3.5 inches. Therefore, row tear-out
capacity is calculated as:
ZRT’ = nrow ni Fv t scritical
(2)(8)(150)(3 5)(3 5) = 29,400
29 400 lbs > 20
20,000
000 OK
iti l = (2)(8)(150)(3.5)(3.5)
127
Group Tear-out Capacity
Assuming a uniform row spacing, and edge distance of 3.75 inches,
calculate group tear-out capacity as:
ZGT = ZRT + Ft Agroup-net = (29,400)/2 + 625(3.5)[11.25 – 2(3.75) – (0.9375)]
= 20,850 lbs.
Note that Group-net is the net area between the outer rows in the group,
which is why the bolt holes are subtracted out.
The design is still acceptable. We have met all three checks.
128
What happens if row edge distance is decreased to the minimum permissible
of 1.5 D?
Assuming a uniform row spacing and edge distance of 1.31 inches,
calculate group tear-out capacity as:
ZGT = ZRT + Ft Agroup-net = (29,400)/2 + 625(3.5)[11.25 – 2(1.31) – (0.9375)]
= 31,527 lbs
…a dramatic capacity increase!
129
Well then, what happens if we space the rows really close together on the
NDS minimum spacing?
Engineering Calculations
Assuming a uniform row spacing and inter-row distance of 1.31 inches,
calculate group tear
tear-out
out capacity as:
ZGT = ZRT + Ft Agroup-net = (29,400)/2 + 625(3.5)[1.31] = 17,566 lbs
Not good - in fact dangerous! Message: spread out the fasteners!
130
As in the other chapters, adjustment factors unique to mechanical
connections are described here for both ASD and LRFD processes.
131
The Group Action Factor Cg is provided in the NDS for multiple fastener
connections to account for load distribution within the connection. Nominal
lateral design values for split ring connectors, shear plate connectors, or
dowel-type fasteners with D less than or equal to 1” in a row are multiplied
by Cg. There are two ways to determine Cg: tables and calculation.
132
Let’s first review Cg terms. What is a row? Two or more fasteners aligned in
the direction of load. Determining numbers of rows can also be tricky…here
are some diagrams to assist. Using the ratios in the diagrams helps
determine the number of rows.
133
The calculation equation for Cg is shown here. As noted earlier, tabulated
values are still included in the NDS. This equation is available if the designer
does not wish to interpolate tabulated values.
134
The calculation depends to a degree on the load-slip relationship between
the fastener and the holding material(s). The NDS tabulates the load-slip
modulus for various installations as shown here.
135
Here is an example of a calculation for Cg. The problem overview and
material data are shown here for two rows of 1″ diameter bolts spaced 4″
apart in a wood-to-wood double shear splice connection using 2x12’s for
main and side members.
136
A Cg value of 0.669 is calculated based on the parameters given.
137
Tabulated values can be selected for the same problem since criteria fits the
bounds of the tables in the NDS.
138
The steps are explained as shown. The table provides a Cg value of 0.665,
consistent with the value of 0.669 that was calculated earlier.
139
The Group Action Factor does not apply to sill plates because unit loads are
not necessarily axial along the plate.
140
Another adjustment factor that is important to connections is the wet service
factor, CM.
141
Connection strength varies with wood moisture content (MC), and the NDS
has provisions to this effect - the Wet Service Factor, CM, that adjusts
connection design values. Two conditions of MC at fabrication and in-service
are important: <19% and >19%. The latter condition includes both
continuous or occasional exposure at moisture levels greater than 19%. The
designer must assess the environmental situation to see which occurs when.
At MC levels above 19%, wood is more elastic, and wood strength properties
reduce somewhat. When wood connections are fabricated using wood with
high MC’s over 19%, and MC levels are expected to drop to final values
below 19% in service, considerable shrinkage takes place around the
fasteners, and grouped fasteners are especially vulnerable to tension
perpendicular to grain stresses; hence the low value of CM = 0.4 for lateral
load conditions where D>1/4”
D>1/4 .
142
The NDS has a detailing provision for dowel-type connections with D>1/4”
that can provide a wet service factor equal to 1.0.
CM=1.0 for dowel-type fastener connections with:
- one fastener only, or
- two or more fasteners placed in a single row parallel to grain, or
- fasteners placed in two or more rows parallel to grain with separate splice
plates for each row.
Minimum distances between fasteners and between fasteners and edges
still need to be maintained. This detailing allows wood to shrink across the
grain without being resisted by the fasteners - the fasteners can move with
the wood.
143
Keep spacing between rows of bolts on a common splice plate to less than 5
inches to avoid splitting the wood due to changes in moisture content.
144
Chapter 11 deals with dowel-type fasteners. Table 10.3.1 contains ASD and
LRFD adjustment factors for various fastener types and loading directions as
described earlier. Chapter 11 specifically covers bolts, lag screws, wood
screws, and nails and spikes.
145
This table shows which fastener types are covered in the NDS versus those
covered by national evaluation reports also called evaluation service reports.
Evaluation reports are developed for proprietary products and provide
designers and code officials with the appropriate information to design
fasteners per the NDS. However, if designing fasteners tabulated in the
NDS, bolts, lag screws, and wood screws must conform to the applicable
ANSI/ASME Standard referenced for these fasteners in 11.1.2, 11.1.3, and
11 1 4; and nails and spikes must meet the ASTM requirements specified in
11.1.4;
11.1.5.
146
Design values for connections tabulated in NDS Chapter 11 are based on
several properties including fastener bending yield strengths, Fyb, given in
footnotes of the respective tables. Other fastener bending yield strengths
may be used with yield mode equations to calculate design values for nontabulated fasteners provided the designer has the appropriate properties.
147
Bending yield strengths, Fyyb, of nails and spikes may be determined in
accordance with ASTM F1575-95 (see Appendix I of the NDS) in lieu of
tabulated bending yield strengths.
148
Another required property for the yield limit equations is the dowel bearing
strength, Fe, of fasteners which is calculated for lumber based on specific
gravity, G, and dowel diameter, D. Fe is tabulated for plywood and OSB.
Structural composite lumber manufacturers list equivalent lumber G values
in their evaluation reports to determine Fe for their proprietary products.
149
There are four possible yield modes for dowel-type fasteners. Yield
equations for connections in single and double shear are included. Wood-towood, wood-to-steel, and wood-to-concrete connections are also tabulated.
150
Reduction terms, appearing in the denominator of the NDS yield equations,
vary by dowel type. To facilitate a general format for the yield limit equations,
reduction terms have been separated from the yield equations as shown
here.
151
There are many variations of a nail as shown here, with a variety of names,
even variations in the way they are installed. Nail capacities are tabulated for
only some of them, such as box and common nails since these are
standardized in ASTM F 1667. The NDS equations can also be used to
develop design values for other types of nails if required material properties
are procured by the designer.
152
Appendix L of the 2005 NDS describes and details dimensions for common,
box, and sinker nails.
153
In terms of shank diameter, same-designation box, common, and sinker nails
are NOT necessarily the same: a 6D common is similar to an 8D box, for
example. Shank diameters differ among same-designation nail types. This
table is an excerpt from an NDS nail capacity table that shows side-by-side
designations of common, box, and sinker nails based on shank diameter.
One important factor in nail capacity determination is nail shank diameter as
seen in capacity formulas on which the table is based.
154
The 1997 NDS (and 1996 LRFD Manual) only required the designer to check 3 yield mode
equations for wood screw and lag screw connections or 4 yield mode equations for nail
connections in single shear. The penetration depth factor, Cd, was assumed to account for
the other modes.
The 2001 NDS eliminated the penetration depth factor for nails, wood screws, and lag
screws. The removal of this factor was coupled with the requirement to check all yield limit
equations per section 11.3.1. This change allows the effect of reduced penetration on
strength to be calculated in a consistent manner with the yield mode equations. Nails in
double shear now need to be calculated using the double shear equations.
The NDS still has provisions for the minimum penetration permitted. For lag screws, this
penetration limit is 4 fastener diameters (D) excluding the tip. For nails and screws, this
penetration limit is 6D including the tip, except in cases where 12D or smaller nails are used
in double shear. When this exception occurs, the side member must be at least 3/8" thick
and the nails must extend at least 3D beyond the side member and be clinched.
Tabulated lag screw, wood screw, and nail values were calculated using penetrations of 8D,
10D, and 10D respectively. For users that rely on tabulated values for design rather than the
calculation method, values for connections with reduced penetration can be conservatively
calculated using the table footnotes. Note that main member thickness is assumed to be
sufficient to provide full penetration of the fastener, except where noted in the table
footnotes.
All of these changes carried forward to the 2005 NDS. A connection calculator is available
on the AWC website which provides a way to calculate design values for dowel-type
connectors: http://www.awc.org/calculators/connections/ccstyle.asp
155
To be effective in developing its full capacity, fasteners must achieve a minimum penetration
d th iinto
depth
t th
the main
i member
b as iindicated
di t d iin th
the ttable.
bl IIn th
the 2001 NDS,
NDS the
th minimum
i i
penetration for wood screws was increased to 6D from 4D.
When the penetration depth factor adjustment was removed in the 2001 NDS, it was felt that
penetrations used to calculate yield strength should all be consistent. Technically, the
preferable way of dealing with it would be to exclude a blunt tip (like a lag screw) or take
some average tip length for tapered tips (like a common nail), but there is no standard on tip
configurations for nails and wood screws.
screws However,
However tip lengths for diamond
diamond-point
point nails
nails, such
as common and box nails, range from 1.3D to 2D. Lag screw tip lengths are shown in
Appendix L.
After evaluating yield limit equations, it was determined that an effective "tip length" of 2D in
the dowel bearing length does not significantly impact the estimated fastener capacity when
fastener penetration exceeds 10D.
156
Minimum penetration into the main member (holding member) for nails is 6
nail shank diameters (D) measured from the nail tip.
157
Minimum penetration into the main member (holding member) for wood
screws is 6 screw shank diameters (D) measured from the screw tip. This
applies to both cut thread and rolled thread screws.
158
Lag screw penetration is different. Minimum penetration into the main
member (holding member) for lag screws is 4 screw shank diameters (D)
measured from the distance E (tabulated in Appendix L) from the screw tip.
The screw tip is not included in the measurement. This applies to both
reduced body diameter and full body diameter lag screws.
159
For reduced body diameter lag screw lateral capacity, the thread root
diameter, Dr, is used in the calculation and tabulated values in the 2005
NDS. Dr is used no matter where the shear plane is located along the length
of the lag screw.
160
Full body diameter lag screws are different. Conservatively, the thread root
diameter, Dr, is used to determine calculated and tabulated values of lateral
capacity in the 2005 NDS. However, the shank diameter can be used
provided the shear plane is located sufficiently away from the threads. Per
NDS 11.3.6.2, the bearing length of threads cannot exceed ¼ the total
bearing length in the main member. This is to enable threads of bolted
connections to be just inside the outer face of the main member to permit
tightening of the nut without having threads in proximity of the shear plane
plane.
161
Even if the shear plane is located in the shank of the screw, the point of
maximum moment actually occurs in the threads well back towards the
screw tip. This is where Dr governs. It is not until the shear plane gets far
enough away from the threads that maximum moment can develop in the
screw shank, permitting the use of the shank diameter, D, in the lateral
capacity determination. For this calculation, and more information on where
to locate the shear plane to permit the use of D, refer to TR12 General
Dowel Equations for Calculating Lateral Connection Values.
162
For this calculation, and more information on where to locate the shear plane
to permit the use of D, refer to TR12 General Dowel Equations for
Calculating Lateral Connection Values, and DA1.
163
As for reduced body diameter lag screw lateral capacity, the thread root
diameter, Dr, is used in the calculation and tabulated values in the 2005 NDS
for rolled thread wood screws. Dr is used no matter where the shear plane is
located along the length of the wood screw.
164
For full body diameter lag screws, the same holds as for cut thread wood
screws. Conservatively, the thread root diameter, Dr, is used to determine
calculated and tabulated values of lateral capacity in the 2005 NDS.
However, the shank diameter can be used provided the shear plane is
located sufficiently away from the threads, about 3 or 4 shank diameters.
Again, see TR12 for details.
165
On toe-nailing, the NDS provides the following guidance:
11.1.5.4 Toe-nails shall be driven at an angle of approximately 30° with the
member and started approximately 1/3 the length of the nail from the
member end.
11.5.4.1
11
5 4 1 When toe
toe-nailed
nailed connections are used
used, reference withdrawal design
values, W, for the nails or spikes shall be multiplied by the toe-nail factor, Ctn
= 0.67. The wet service factor, CM, shall not apply for toe-nailed connections
loaded in withdrawal.
11.5.4.2 When toe-nailed connections are used, reference lateral design
values, Z, shall be multiplied by the toe-nail factor, Ctn = 0.83.
166
Chapter 11 provides definitions and variables leading to determination of
capacity. One of the most important is direction of applied load with respect
to the grain of the wood. From the load path, load direction with respect to
wood grain can be determined for each wood component in the connection.
This helps with selection of the correct design value from the NDS fastener
capacity tables.
167
The NDS does contain minimum spacing, edge, and end distance rules for
fastener placement. Again, load direction can play a role in their
determination. Correct fastener placement to develop full design capacity of
the fastener may also be governed by the provisions of Appendix E on
checking local stresses.
168
As a connection example with the 2005 NDS, consider a nailed shear wall
chord tie design first by using 2005 NDS tables for both LRFD and ASD.
Then compare results.
169
The first practical consideration in this case is to choose a fastener type.
Many proprietary pre-fabricated metal connectors are available to make this
connection. However, a connection can be designed that will use commonly
available, non-proprietary, components.
170
Material design parameters are assumed for the first trial.
171
First, determine the unfactored unit capacity, Z, of the nail (ASD value) from
2005 NDS Table 11P.
172
Now determine applicable adjustment factors from 2005 NDS Table 10.3.1.
Conversion factors for LRFD are obtained from Appendix N.
173
Factor the loads for LRFD, noting that this is for wind uplift only (no dead
load for this example).
174
Determine the LRFD and ASD capacities accordingly for one nail, then
dividing into the demand, determine the number of nails required. Note the
difference in results between the ASD and LRFD methods.
175
Taking a ratio of the results for ASD and LRFD on the basis of demand over
capacity reveals that LRFD is more conservative than ASD. Why the
discrepancy?
176
The format conversion from ASD to LRFD in the 2005 NDS does not benefit
LRFD for the wind-only case. However, real benefits are realized with
combined multiple transient live loads (e.g. wind + snow + live). Examining
load combinations and load factors in addition to relative magnitudes of the
loads themselves, reveals a more realistic assessment of LRFD versus
ASD, since ASD demand is usually a straight summation of load.
177
AWC’s Connection Calculator provides users with a web-based approach for
calculating capacities for single bolts, nails, lag screws, and wood screws
per the 2005 NDS. Both lateral (single and double shear) and withdrawal
capacities can be determined. Wood-to-wood, wood-to-concrete, and woodto-steel connections are possible. ASD or LRFD approach.
178
Chapter 12 features information on split ring and shear plate connectors.
ASD capacity tables have not changed for many editions of the NDS, and
this is still true for the 2005 edition. These devices are high capacity
fasteners meant for use in very large members and member cross-sections.
179
Chapter 13 is for timber rivets, a very useful and effective device for
connecting members of small or large cross-sections. The capacity tables
remain unchanged from the 2001 NDS.
180
Timber rivet connections have been used in Canada for several decades.
The design criteria introduced in Chapter 13 of the NDS apply to joints with
steel side plates for either Southern Pine or Western Species glued
laminated timber. The term "timber rivet" was chosen to accommodate future
application to sawn lumber as well.
Provisions of the Specification are applicable only to timber rivets that are
hot-dipped galvanized. Rivets are made with fixed shank cross-section and
head dimensions (Appendix M) and vary only by length.
Because of the species test results and property values used to develop the
rivet bending and wood capacity equations, use of design values based on
provisions of 13.2.2 should be limited to Douglas
g
fir-Larch and southern
the p
pine glued laminated timber manufactured in accordance with ANSI/AITC
A190.1. The NDS presently limits use of timber rivets to attachment of steel
side plates to glued laminated timber.
181
The 2005 NDS specifies timber rivets made of mild steel (AISI 1035), and
plates of A36 steel. Further, design provisions and values of the 2005 NDS
are applicable only to timber rivets that are hot-dipped galvanized. Plates
also need to be hot-dipped galvanized if the connection is in wet service.
This is all described in 2005 NDS 13.1.1.
Good practice is to always hot-dip galvanize metal components for corrosive
or exposed environments and in situations where the structure may be
exposed to the elements for long construction periods that might result in
streaking stains on the wood that can be very difficult to remove (unsightly if
the final structure is meant to be exposed for aesthetics).
182
Rivets are made with fixed shank cross-section and head dimensions
(Appendix M) and vary only by length.
183
Plates also have a fixed hole pattern geometry. Hole sizes are chosen
deliberately to firmly hold and lock the head of the rivet in position,
preventing the rivet from rotating next to the plate, to fully develop a
cantilever action for the rivet shank embedded in the wood. Note the
minimum edge distances and hole spacings – these are used to enter the
capacity charts in the NDS.
184
Rivet connections can be made from one or both sides of a member.
185
Wood members can be loaded parallel or perpendicular to grain. However,
note that the major cross-sectional dimension of the rivet shall be aligned
parallel to grain.
186
Angle to grain capacity values are also provided in the NDS. Note that rows
always align parallel with the direction of loading on the plate.
187
Essentially, there are four strength limit states for a timber rivet connection;
two parallel-to-wood-grain (P-direction), and two perpendicular-to-wood grain
(Q-direction). For each grain direction, either the rivet yields or the wood
fiber yields. If the load is applied only in the P-direction, or only in the Qdirection, then the number of strength limit states to check reduces to two:
rivet yielding and wood capacity. The lower capacity will govern the design.
The perforated plate is stiff and, although rarely an issue, should also be
checked using appropriate steel code provisions.
188
The design process is simple, regardless whether ASD or LRFD is used, and
is best implemented using a spreadsheet, or other calculation software
because of its iterative nature. Here are the primary features of the
connection that the designer can vary to get the required capacity with
reasonable ductility.
189
Here are the design steps. After determining the total loads that must be
resisted (demand), assume a trial design based on connection configuration
geometry that will accommodate a grid of rivets, minding tabulated minimum
edge and end distances. The main variables here are: plate thickness, rivet
length, rivet spacing parallel to wood grain, number of rows of rivets, and
number of rivets in each row.
190
Next, check rivet yield – an equation is given for this based on the capacity
of a single rivet through a single plate. There are two equations: one each
for the P and Q directions respectively (NDS 13.2-1 and 13.2-2).
191
Now, check wood capacity parallel-to-grain (P-direction) – from a table
based on rivets installed on faces of the connection. The tables are
organized by rivet length and by plate thickness for typical rivet grid
spacings. Footnotes to the table offer explanations of member width. The
tables simplify the design process tremendously and allow the designer to
avoid using complex equations for predicting wood capacity in shear or
tension. The equations were originally developed and verified by tests. For
details on these equations
equations, see the 2005 NDS Commentary.
NDS Tables 13.2.1A – 13.2.1F are for connections with steel side plates on
opposite sides of the wood member. The reference design value in the table
is for the capacity of one ¼” side plate with associated rivets (NDS C13.2.1).
Thus for a connection with plates on opposing faces, the designer would
double the table value to determine the reference capacity of the connection
connection.
For connections with a single plate of rivets on one side of the wood
member, the designer enters the table with twice the thickness of the wood
member to get the correct reference capacity for a single-sided connection.
192
Check wood capacity perpendicular-to-grain (Q-direction) – an equation
(NDS 13.2-3) is given for this based on the capacity of a single rivet through
a single plate. The equation references two tables as shown.
193
One table is for the reference value (NDS Table 13.2.2A) based on one plate
with rivets installed in one side of the connection.
194
Another table is for the Geometry Factor, C , (NDS Table 13.2.2B). Again,
the reference design value obtained from the equation is doubled for
connections having two side plates.
195
The lowest capacity of four checks above will govern the capacity of the
connection. If rivet yield governs, then greater ductility of the connection is
assumed. If wood capacity controls, the the connection is likely to be less
ductile.
196
Adjust the determined capacity for site environmental conditions using
adjustment factors from Table 10.3.1, minding the applicability notes at the
bottom of the table.
One of the more important notes is Footnote 4 in the ASD application of CD
when rivet yield governs the strength design. Note that CD drops out of the
ASD capacity when rivet yield controls (Footnote 4), yet λ remains on the
LRFD side. For LRFD, the time effect factor, λ, applies to Pr and Qr since
the format conversion factor, KF, for connections adjusts from a 10-year to a
10-minute load basis. CD does not apply for ASD values of Pr and Qr
(Footnote 4) because "rivet bending capacity" was treated as a steel limit
state in early research and implementation. The early assumption was that
rivet bending capacity is unaffected by load duration. Load duration effects
were specifically considered in checks of wood strength limit states
states, not steel
strength limit states.
197
Another example - the adjustment for wet service conditions…
198
… and for elevated temperature …
199
… and for metal side plates of various thicknesses.
200
Here is how the adjustment factors are implemented in the directional
capacity equations.
201
Finally, determine the governing capacity (minimum value) …
202
…and then calculate the demand:capacity ratio – a value less than 1.0 is
OK.
203
If the ratio is greater than 1.0, try adding more rivets and repeat the trial
design. If the number of rivets is not in the table, try increasing the rivet
spacing parallel-to-grain and move to another table. Still no good? Try
increasing the plate thickness. Still not enough? Try increasing the rivet
length in increments to the maximum penetration permitted by the
connection geometry, and repeat the trial.
204
The design process for timber rivets will be illustrated through two examples
of typical connections using timber rivets. Each example is worked in both
ASD and LRFD and is based on the 2005 NDS timber rivet provisions found
in Chapter 13.
Each solution has been developed using Mathcad® software by Parametric
Technology Corporation® (PTC®). Therefore, formatting of certain variables
and equations as shown in the examples are unique to this software.
205
For the first example, consider a simple tension splice loaded in the wood
parallel-to-grain direction (P), with rivet plates installed on opposing wood
faces. Here, three strength limit states are of interest: rivet strength, parallelto-grain wood strength, and tensile strength of the perforated connecting
plate.
206
Connection geometry.
207
Note that C does not apply here because of the loading direction. See
Table 10.3.1 Footnote 6.
208
Begin with factored and unfactored loads.
209
Determine wood and rivet capacities. Looks like wood has the lower number.
210
Since wood controls, the adjustment factor Cst is not used. Adjusting the
controlling result, and doing the demand:capacity comparison gives a
satisfactory answer. The ratios are a little low, so the design could be
reworked reducing rivets to provide a more economical result.
211
Graphic of the solution.
212
Now, check the plate. There are three checks on the plate required.
213
The first is gross area yielding.
214
Second, net section through the plate perforations.
215
Lastly, block shear fracture. It’s interesting to note that these steel plate
checks are similar to the wood fastener provisions of NDS Appendix E.
Checking rivet group block pull-out failure of the wood member, or other
known local stress effects due to the rivets, is not needed since these failure
modes were included in the generation of the 2005 NDS timber rivet table
values – see 2005 NDS E.1.1 for information.
216
For the second example, consider a beam-to-girder hanger connection, with
the hanger installed with rivets to one wood face of the girder and loaded in
the girder perpendicular-to-grain direction. Here, two strength limit states
are of interest: rivet strength, and perpendicular-to-grain wood strength. The
hanger is assumed to be structurally adequate. In the example, three trials
are run. The first trial with the wood capacity governing does not work,
however the second trial where rivet capacity governs does work simply by
adding more rivets and providing desirable connection ductility
ductility. The third
trial shows a way of preserving the desired rivet yielding mode with fewer
rivets, by relocating the rivet array closer to the top face of the girder.
217
Here’s the geometry. Note in the elevation how the rivets are spread out
vertically on the plate next to the girder.
218
Treat the hanger face as two plates separated by a carried beam.
219
Start by calculating factored and unfactored loads.
220
Determine wood capacity perpendicular to grain.
221
Determine wood capacity perpendicular to grain (continued).
222
Determine rivet capacity.
223
Wood capacity controls, and the trial is no good.
224
In the next trial, increase the number of rivets in each row by two.
225
This time, rivets have the lower capacity.
226
Since rivets control, pay particular attention to the application of CD in
Footnote 4 of Table 10.3.1, which only affects the ASD capacity as
explained in the slide.
227
Applying the adjustments correctly, this trial works. Looking at the
demand:capacity ratio for the LRFD result shows a slight advantage over
ASD where we could optimize by using fewer rivets in the LRFD design.
228
Here’s the tentative solution from Trial 2. This can be optimized further.
229
See the refinement notes in the slide. Here ep is minimized to max out the
shear coefficient of the girder.
230
Reworking the design.
231
Use 10 rivets in each row per Trial 2.
232
Setting ep to 1 inch really increases the value of C (almost triple).
233
This really increases the wood capacity, causing the rivet capacity to control
the design.
234
The 10 rivets per row solution works, just by relocating them higher up on
the girder face. Working the LRFD design using 8 rivets per row might also
work.
235
Here’s the final solution.
236
American Forest & Paper Association / American Wood Council
APA - The Engineered Wood Association
Wood Truss Council of America
Canadian Wood Council
A more comprehensive program on connection design is available on the
AWC website.
Copyright
© 2005-2011 American Wood Council
Copyright © 2001, 2007 American Forest & Paper Association Inc., APA - The Engineered Wood Association,
Wood Truss
Council of America Inc., Canadian Wood Council, Inc. All rights reserved. For permission
AllInc.,rights
reserved.
to reprint contact AF&PA at 1-800 AWC-AFPA.
237
Chapter 14 begins the sections of the NDS dealing with special provisions.
Chapter 14 on shear walls and diaphragms covers general requirements for
framing members, fasteners, and sheathing. The reference document for the
design process of shear walls and diaphragms is AWC’s Special Design
Provisions for Wind and Seismic standard.
238
The Special Design Provisions for Wind and Seismic standard, known as the
Wind and Seismic standard, is the scope of another course. In addition to
design process for shear wall and diaphragm elements, the Wind and
Seismic standard includes reference design values for a wide variety of
panel products, as well as a Commentary to the provisions. The table of
contents of the document is shown here.
239
Chapter 15 on Special Loading describes various topics related to loads
such as: lateral distribution of a concentrated load, spaced columns, built-up
columns, and wood columns with side loads and eccentricity.
The 2005 NDS revises a limitation on short built-up columns whereby the
designer can use the lesser of the column capacity reduced on the basis of
slenderness of the entire cross-section, and the column capacity of an
individual lamination multiplied by the number of laminations.
240
Chapter 16 on the design of exposed wood members to meet building code
prescribed fire endurance times first introduced in the 2001 NDS is only
applicable to ASD design.
241
ASD provisions address tension, compression and bending members and
members subjected to combined loading. Special provisions for glued
laminated timber beams are also included.
242
The basis for Chapter 16 is found in AWC’s document TR 10: Design of Fire
Resistive Exposed Wood Members
This document also forms the technical basis for AWC’s DCA 2. It is
complete with detailed explanation, test results, and comprehensive
calculation examples.
243
The physical basis for Chapter 16 is the charring characteristic of wood
when subjected to fire. Charring of wood occurs at a measurable rate, and
because of wood’s insulation properties, the cross-section interior remains
capable of sustaining and carrying load.
244
Charring rates of wood under standard fire exposure conditions were
measured in studies world-wide. Glued products did not perform any
differently than their solid counterparts.
245
This design method is a rational approach that allows for exposed structural
wood members to be used in structures that could be exposed to fire.
246
The equations used in this method account for all the charring characteristics
of a wood cross-section exposed to fire.
247
A standard terminology was established for describing the charred and uncharred
section dimensions for
f two common fire
f exposures.
248
…which resulted in these relations for charred width and depth.
249
In terms of the charring characteristics of wood, this is the char model used.
250
…and these are the charring results based on a typical char rate of 1.5
inches per hour.
251
The factor, K, adjusts from allowable design capacity of the member to
average ultimate capacity - the maximum capacity the member can
physically sustain (no safety factors).
252
This table lists the values of K for various capacities to adjust to an ultimate
strength basis.
253
Given the theoretical derivation of the new mechanics-based design method,
existing test results from fire tests of exposed, large wood members were
compared against the model predictions and were found to be excellent
agreement. Here is one such example where the model and test agreement
were good for wood beams exposed on 3 sides.
254
ASD example.
Consider Douglas fir beams spanning 18 feet and spaced 6 feet apart. The
beams support 100 psf live load and 15 psf dead load. Timber decking
laterally braces the compression flange of the beams.
Size the beam for a 1 hour rating
rating.
255
Solution:
First, calculate the induced demand moment based on the tributary width of
6 feet (beam spacing).
256
Select a trial beam, calculate its section modulus from actual dimensions,
and the adjusted allowable bending stress of the material.
257
Multiply the adjusted allowable bending stress by the section modulus to get
the maximum resisting moment offered by your chosen beam. Check for
adequacy, and in this case, OK.
258
Now, design the cross-section for fire endurance. A certain amount of the
cross-section will char during the duration of the rating time, reducing the
cross-section size required to sustain load.
From the table in Chapter 16, find the char depth for the duration you are
seeking, in this case, 1 hour.
259
Determine the charred section dimensions and calculate a new charred
section modulus for the residual section.
260
Recalculate the adjusted allowable bending stress, since not all of the
adjustment factors apply here and may have been a value other than 1.0
before.
Determine the strength resisting moment based on the charred crosssection, and in this case is good for a 1 hour fire duration.
261
The modeled behavior is conservatively accurate, can be easily
implemented as a design process, and permits designers to use exposed
large section wood members in structural applications that could be subject
to fire exposure.
262
2005 NDS Appendix E has remained substantially the same. Appendices N
is the only new one, and is a mandatory part of the standard necessary to
provide the LRFD element to the NDS.
263
Appendix N provides the necessary tables for LRFD implementation. ASCE
7-02 is the reference load document. However, footnote 2 of NDS Table N3
provides clarification that the specific load factors shown are for reference
only and are intended to provide flexibility in assignment of the time effect
factor in the event of changes to specified load factors.
The time effect factor,  (LRFD counterpart to the ASD load duration
factor, CD), varies by load combination and is intended to establish a
consistent target reliability index for load scenarios represented by
applicable load combinations. With the exception of the load combination for
dead load only, each load combination can be viewed as addressing load
scenarios involving peak values of one or more “primary” loads in
combination with other transient loads. Specific time effect factors for various
ASCE 7 load combinations are largely dependent on the magnitude
magnitude,
duration, and variation of the primary load in each combination. For example,
a time effect factor of 0.8 is associated with the load combination 1.2D + 1.6
(Lr or S or R) + (L or 0.8W) to account for the duration and variation of the
primary loads in that combination (roof live, snow, or rain water, or ice
loads). The effect of transient loads in a particular load combination or even
changes in the load factors within a given combination is considered to be
small relative to the effect of the primary load on the load duration response
of the wood. Consequently, specific time effect factors need not change to
address load factor or load combination changes over time.
264
The 2005 NDS Supplement contains all of the reference design values for
various lumber and certain engineered wood products, and is part of the
standard.
265
A new feature of the NDS Supplement that corresponds to a change in NDS
provisions is the tabulation of the 5th-percentile E values used in beam
stability and column design equations. Emin translates well between the ASD
and LRFD processes through the tabulation. Thus, reference design value
tables for all lumber and engineered wood products now include the Emin
values in their tables.
266
Reference design value data has been added for four new wood species of
lumber…
267
… as well as two new species of timber.
268
The list of non-North American Species continues to grow, adding several
new species to the list of tabulated reference design data.
269
New design values have been added for mechanically graded dimension
lumber. Specifically, footnote 2 of Table 4C in the NDS Supplement provides
specific gravity, shear parallel to grain, and compression perpendicular to
grain design values for machine stress rated (MSR) and mechanically
evaluated lumber (MEL). Table 2 provides an overview of the new design
values for MSR and MEL lumber. As with visually graded lumber and
timbers, modulus of elasticity for beam and column stability, Emin, design
values have been added to Table 4C for MSR and MEL lumber
lumber.
270
Several changes have been made to structural glued laminated timber design values in the 2005 NDS
Supplement. As with dimension lumber and timber tables, modulus of elasticityy for beam and column stability,
y
Emin, design values have been added for glued laminated timber. Species groups for split ring and shear plate
connectors were removed from Tables 5A–5D. In some cases, these groups did not correspond to species groups
assigned according to NDS Table 12A. A review of the data used to establish connector species groups indicated
that values in Table 12A are appropriate. Specific gravity, G, of the wood located on the face receiving the
connector should be used with NDS Table 12A for assignment of species group. This change is consistent with
current recommendations of the American Institute of Timber Construction (AITC) and APA–The Engineered
Wood Association.
There were specific changes to Tables 5A, 5A-Expanded, and 5B. Design values for tension parallel to grain, Ft,
compression
parallel to grain, Fc, and specific gravity, G, are revised for the 16F stress class. The 2001 NDS Supplement
showed
h
d diff
differentt values
l
ffor thi
this stress
t
class
l
iin T
Table
bl 5A vs. 5A
5A-Expanded.
E
d d A
Analysis
l i iindicated
di t d th
thatt th
the values
l
iin
Table 5A-Expanded were correct, so Table 5A was updated accordingly.
Shear parallel to grain (horizontal shear) design values have increased for prismatic members, and adjustment
factors in accordance with Footnote d have been revised. Horizontal shear values in the 2001 NDS Supplement
were based on full-scale tests of laminated beams, which were reduced by 10 percent based on judgments made
at that time. Shear values for non-prismatic members were those derived according
to ASTM D3737 from tests of small shear-block specimens. Since that time, the structural glued laminated timber
industry has revised its recommendations and has elected to publish test-based shear values for prismatic
members, removing the 10 percent reduction. This change
is reflected in the 2005 NDS Supplement
pp
consistent with recommendations of AITC and APA. Footnote d
adjustment factors were revised to keep shear values for non-prismatic members essentially unchanged.
Historically, radial tension design values for structural glued laminated timber were established as one-third of
shear parallel to grain design values. In the 1991 NDS, radial tension values were 67 psi for Southern Pine and 55
psi for Douglas Fir-Larch, respectively. For Douglas Fir-Larch, radial reinforcement designed to carry all induced
stresses was required to utilize this value, otherwise the radial tension value was limited to 15 psi–this point was
clarified in the 2005 NDS. Comparing 2005 to 1991 NDS Supplements, increased shear values for non-prismatic
members of Douglas Fir-Larch and Southern Pine have resulted in small increases for radial tension design
values in these species. The slightly increased radial stresses are recommended by AITC and APA and are
considered appropriate and preferable to multiple adjustment factors as were used in the 2001 NDS.
271
Table 5B of the NDS Supplement incorporates the following changes:
• Re-formatting of bending design values for bending about the X-X axis, Fbx.
If special tension laminations are included, tabulated values may be adjusted
according to applicable footnotes.
• New combinations for Southern Pine were added with extra information
regarding slope of grain differences.
Y Y axis,
axis
• Shear value columns were consolidated for bending about the Y-Y
Fvy, and shear values were updated consistent with Table 5A discussion
above.
The most notable change to all design value tables in the NDS Supplement
is the addition of minimum modulus of elasticity values for beam and column
stability Emin
stability,
design The change to shear design values for prismatic glued
i , design.
laminated timber members is another significant modification.
272
For 2005, the NDS provides a new format for the future that allows two
design processes to be used: ASD, and LRFD. Further, the new NDS binds
in one volume: provisions, design values, and commentary.
273
The complete wood design package adds three more documents: the Wind
& Seismic standard for lateral design of wood structures, the Manual filled
with helpful non-mandatory information in the application of the NDS to wood
building design, and an Examples workbook of ASD and LRFD practical
design examples to shorten the learning curve.
274
The ANSI/AF&PA SDPWS-2005 covers materials, design and construction
of wood members, fasteners, and assemblies to resist wind and seismic
forces. Engineered design of wood structures to resist wind or seismic forces
is either by allowable stress design (ASD) or load and resistance factor
design (LRFD).
275
The ASD/LRFD Manual contains design information for structural lumber,
glued laminated timber, structural-use panels, shear walls and diaphragms,
poles and piles, I-joists, structural composite lumber, metal plate connected
wood trusses, and pre-engineered metal connectors. Over 40 details are
included in the chapter on connections. A comprehensive chapter on fire
design includes fire rated wall and floor assemblies for solid sawn lumber, Ijoists, and trusses.
276
Structural Wood Design Solved Example Problems is intended to aid
instruction on structural design of wood structures using both allowable
stress design and load and resistance factor design. Forty example
problems allow direct side-by-side comparison of ASD and LRFD for wood
structures.
277
Here is a summary list of all the changes to the 2005 NDS document from
the 2001 version….
278
…and another for the 2005 NDS Supplement.
279
Together, the four volumes form the 2005 Wood Design Package.
280

American Wood Council Welcome to STD 104: ASD and LRFD with

Transcription

Similar documents

Warranty Info

Hertco Express 20-DoorStyles_062013-A.indd

Riddles taken from I Am a Rock

1923 Ford Model T Huckster for Sale

Materials Matter - Fort Street Studio

PDF

www.FirstColorado.com 18205 Highway 285 Nathrop, CO 81236

Terra Plus Technical Data Sheet

Premium Wood Stain Brochure - Rust

BANGKI-OIL