Copyright © 1999, 2004 by American Forest & Paper Association, Inc., Washington, D.C.

ADS Commentary

_-
PART IX: LAGSCREWS

9.1-GENERAL cation sincethe1944 edition. The leadholerequire-

ments for the three specific gravity classes are based on
9.1.1-Quality of Lag Screws earlylagscrewresearchinvolvingtestsofDouglas-fir,
Methodology given in the Specification for the southern pine,white oak, redwood and northern white
establishmentofdesignvalues for lagscrews prior to pine (1 34).
the 1957 edition wasreferenced to lagscrewshaving 9.1.2.2 The provision for allowing 318 inch and
a yield strength of45,000psi and a tensile strength of smaller diameter lagscrews loaded primarily inwith-
77,000psi.Where lag screwshaving other strength drawal to be insertedinwoodofmediumtolow
propertieswere used, withdrawaldesignvalues and specificgravitywas added to the Specificationinthe
lateral design values were to be adjusted in proportion 1982 edition. Consideration of this limited exception to
to the ratio of tensile strength and in proportion to the theleadholerequirementwas initiated byfield inqui-
ratio of the square root of theyield strength, respec- ries concerning the acceptability of using power driven
tively. In the 1957 edition these adjustment provisions, toolstoinsertsmalllagscrews to join membersin
based on early lag screw research (134), were continued factory builthousing construction. Onthebasis of
but the reference to specific strength metals was early lag screw research (134), available information on
replacedwithreference to commonlagscrews. thewithdrawalresistance of tapping screwsinserted

edition. toThe conditions

exception
qualifying
and related

strength properties. In the 1991 edition, the designer A lag screw subjected to both combined withdrawal
isresponsible for specifyingthemetal strength of the and lateral loading may be considered loaded primarily
lagscrews that are to be used. Bendingyield strength in withdrawal when the axis of the screw is at angle of
of the lagscrewis a required input variable to the 75" or more to the grain of the wood member holding
lateral designvalueyield mode equations of 9.3.1. the threaded portion ofthescrew. The requirement
Additionally, the actual tensile stress in thelagscrew that unusual splitting be avoided whenleadholes are
at the root diameter must be checkedwhendesigning not used is to be considered a performance requirement
lag
screw connections for withdrawal (see
9.2.3
of that (i) is related to the ability of the screw to hold the
Specification). cleat or side member to the main or foundation
9.1 -2-Fabrication and Assembly member, and (ii)is applicable to both membersbeing
c
joined.
9.1.2.1 Provisions relating to the clearance hole of
the shank and the leadholesize for thethreaded 9.1.2.3 The provision that lagscrewsbeinserted
portion of the lag screw have been part of the Specifi- by turning with a wrench and not by drivingwith a

Lag Screws 115

NDS Commentary

hammer has beenagoodpracticerequirementofthe grainofthemainmember,thelengthofthetapered

Specificationsincethe1944edition. tipofthescrewis notto beincluded.Thistapered
portion at thetip of thescrewwas not considered as
9.1.2.4 Use ofalubricant to facilitatelagscrew part oftheeffectivepenetration depth in theoriginal
insertionalso has beenagoodpracticerequirement joint tests(134). In addition, thethicknessofany
since1944. This requirement is not waivedwhensmall washer used betweenthescrewhead and thecleat or
diameterscrews are insertedwithouttheuseoflead side
member should
be
taken into account when
holes. determiningthelengthofpenetrationofthethreaded
9.2-WITHDRAWAL DESIGN VALUES portionin
the
mainmember. Standard lag screw
dimensions, including
thread
length and length of
9.2.1-Withdrawal from Side Grain tapered tip, are giveninAppendix L oftheSpecifica-
tion.
The methodology used to establishthelagscrew
withdrawaldesignvaluesgiveninTable9.2Aisthe 9.2.2-End Grain Factor, CCE
same as that incorporated inthe1944editionofthe
Specification. Tabulated designvalues are computed Tabulated withdrawaldesignvalues for lagscrews
from thefollowing equation based on theresultsof are reduced25percentwhenthescrewisinsertedin
earlyresearch(62,134): theendgrain(radial-tangentialplane)ofthemain
member rather thanthesidegrain(radial-longitudinal
(C9.2-1) or tangential-longitudinal plane).
This
adjustment,
based on lag screw joint tests (134), has been a provi-
where: sion
the
ofSpecification
since
the
1944
edition.
Becauseofthegreaterpossibilityofsplittingwhen
W = withdrawaldesignvalueperinchofpene- subject to lateral load, it has longbeenrecommended
tration into main member, Ibs thatinsertionoflagscrewsinendgrainsurfaces be
K , = 1800 avoided(62,128).
G = specificgravitybased on ovendryweight
and volume 9.2.3-Tensile Strength of Lag Screw
D = lagscrew shank diameter,in. In theoriginallagscrew joint tests(134),penetra-
The value of K , represents one-fifth of the average tion of the threaded portion of the screw into the main
constant at oven dry weight and volume obtained from member ranging from 7 diameters for the highest
ultimate load testsof joints made withfivedifferent densitywood and 11diameters for thelowestdensity
species and seven sizes of lag screw (134), increased by woodtestedwasfoundtodeveloplagscrewstrengths
20 percent; or approximatelyequal to thetensilestrengthofthelag
screw.However,becausetabulatedwithdrawaldesign
K, = (7500/5)1.2 = 1800 (C9.2-2) valuesrepresentonly about 25 percent
of
average
ultimate test loads, the allowable tensile strength of the
The twentypercentadjustmentwasintroduced as screw
associatedwith
the root diameter will not
part of the World War I1 emergency increase in wood generallylimitthewithdrawaldesignvalue.This will
designvalues, and thensubsequentlycodified for lag bethecaseevenwhentabulateddesignvalues are in-
screws as 10 percent for the change from permanent to creased1.6 for wind or earthquake load. (See also
normalloading and 10percent for experience
(see Commentary for 7.2.3 and 9.1.1.)
Commentary for 2.3.2).
9.3-LATERAL DESIGN VALUES
The withdrawaldesignvalueequationabovewas
includedintheearlyeditionsoftheSpecification. 9.3.1-Wood-to-WoodConnections
Beginning withthe1960edition, the equation was Background
replacedbyatableoflagscrewwithdrawaldesign
values for thefullrangeofspeciesspecificgravity Lagscrewsperformthesamegeneralfunction as
values. boltsbut do not requirea nuton the joint face
oppositethefastenerhead. The threadedportion of
It is to benoted that when thetotalallowable thescrewreplacesthe nut inprovidingresistanceto
withdrawaldesignvalue on alagscrewisdetermined withdrawal.Lagscrewstypicallyrangefrom0.19to
by multiplying the tabulated design value by the length 1-1/4inchindiameter and lengthsoflessthan1to
of penetrationofthethreadedportion into theside more than 12inches. The threadedportionofthe

116 Lag Scnws

 NDS Commentary

screwisgenerallyone-half the screwlengthplus 1/2 ratio of 1 to 1.17 for ratio

inch or 6 inches, whichever is shorter (see Appendix L). of 2 to 1.26 for ratio of 3
to 1.39 for ratio of 7)
Lag screw lateral design values provided in the 1986
and all earlier editions of the Specification were estab- F'? = adjustment for depth of penetration inmain
lishedusing the sameprocedures. This methodology member
wasdeveloped from the results of earlylagscrew = DP/RP s 1.0
research (134) whichshowed that proportional limit
designvalues in lag screwconnectionsloadedparallel where:
to grain were a function ofthespecificgravityofthe DP = T-E+SP
members being joined, the shank diameter of the screw, T = length of thread
the ratio of the cleat or side member thickness to the E = length of tapered tip
shank diameter, the location of the shanwthread RP = RSPFIF2 when FIF2 1.0
boundary relative to the edgeofthemain or founda-
tion member, and the depth of penetration of the screw = RSP( FI F2) when FI F2 L 1 .O
in the main member. For perpendicular to grain RSP = required standard penetration
loading, an additional adjustment was introduced for = 7 0 for Group I
area of bearing as related to the diameter of the screw. = 8 0 for Group I1
This adjustment for grain direction was adapted from = 1OD for Group I11
a similar relationship developed for bolted connections = 1 1 0 for Group IV
(62,117,183). The specific formulas used to establish
lag screw values in previous editions are shownbelow: Perpendicular to grain load:
Parallel to grain loading: (C9.3-2)

PI = K'D2 FIF2F.7 (C9.3-1) where:

where: P2 = design load perpendicular to grain

PI = design load parallel to grain
PI = design load, Ibs F4 = adjustment for shank diameter (ranging from
KL = species constant based on specificgravity 1.00 for 3/16 in. to 0.85 for 5/16 in. to 0.70
(G) of wood members for 7/16 in. to 0.50 for 1 in.)
= 2640 Group I G 0.62 - 0.75
= 2280 Group I1 G 0.51 - 0.55 The complexityof the foregoingmethodology
= 2040 Group I11 G 0.42 - 0.49 reflectsthenumber of variables that affectlagscrew
= 1800 Group IV G 0.31 - 0.41 connectionperformance. The equations are basedon
D = shank diameter, in. parallel to grain tests of northern white pine, Douglas-
FI = adjustment for ratio ofcleatthickness ( t ) to fir, southern pine and white oak connections made with
shank diameter (ranging from 0.62 at ratio of dry material (15 percent moisture content) and lag
2 to 1.00 at ratio of 3.5 to 1.22 at ratio of screws having an average tensile strength of 77,000 psi
6.5) (134). The valuesof KL inthe equation for PI repre-
F2 = adjustment for location of shanwthread boun- sent
average proportional limit joint design values
dary divided by 2.25 and thenincreased 20 percent for
normal loading and experience(seeCommentary for
SP= s - ( t + w ) 9.2.1 related to the latter adjustment).

where: In the earlier editions, not all species were assigned

to fastener groups based on their specific gravity alone.
S = lengthof shank Some softwood species having the samespecificgravi-
w = washer thickness ties as species classified in Group I11 wereclassifiedin
= 1/8 in. for D 2 1/2 in. Group IV (62). Theseassignments
paralleled
early
= 0 in. for D < 1/2 in. group assignments for woodscrews and nails.Begin-
ning with the 1971 edition, classification of species into
For SP/D < 0: F2 = 1.0 - 0.2(t+ w - S ) / t
fastener groups for thepurpose of assignment of KL
For SP/D L 0: F2 function of SP/D factors was based
solely on specific
gravity.
The
(ranging from 1.08 for specific gravity classes for KL assignments shown in the

Lag Screws 117

NDS Comentary

legendof Equation C9.3-1were used from the1971 unaffected as long as the total length of penetration
through the 1986 editions. (shank as well as threaded depth) of thescrewinthe
main member was at least five times the shank diame-
The foregoing procedures were included as part of ter, or 5D (1 14,117). In view of the fact that previous
the Specification until the 1960 edition whentheywere designvalueswerekeyed to proportional limit rather
replacedwitha table ofdesignvalues for arange of than ultimate joint loads, the positionwastaken that
lagscrewsizes for each of the four lagscrew groups. the yield mode equations would be implemented in the
This
change in presentation, whichwas made to 1991 edition and compared withpreviouslydeveloped
facilitate design and use of lag screw connections, was designvaluesassuming that the fulldesignvalue is
continued through the 1986 edition. In accordance developed if the length of the screw (shank plus
with recommendations based on the originalresearch threaded portion less length of tapered tip) in the main
(134), tabulated designvaluesin the 1960 through the memberis at least 8D, regardlessofspecies specific
1986 editions were considered directly applicable to lag gravity; and that proportionate design values are
screwshavingayield strength of45,000psi. For lag achieved for penetrations betweenaminimumof 40
screwshaving other yield strengths, adjustment of and 8 0 . These criteria are in agreementwiththe
tabulated values in proportion to the square root of the penetration requirementin the 1988 draft standard of
ratio of the yield strength of the metal to45,000was Eurocode No. 5 (45,114).
allowed.
To establish adjustment factors which would reduce
1991 Edition. As with bolts and other doweltype yield mode equation values to the designvaluelevels
fasteners, lateral design values for lag screw connections used in previous editions of the Specification, the ratios
in the 1991 edition are based on application of the ofyieldmodedesignvalue to designvaluebasedon
yield limit model (see Commentary for 7.2.1 and 8.2.1). previous methodology (modified by the new penetration
In thecase of lagscrews,threegeneralmodesof criteria) were determined for a wide range of lag screw
yielding can occur:bearingin the sidemember or joint configurations made with both wood and steel
cleat (Mode I, ), development of a plastic hinge in the sidemembers. An average ratio was developed for -
screw in the main member (Mode 111) and development each of the seven yield mode conditions, one for Mode
of plastichingesin the screwin both main and side I, and threeeach for Modes I11 and IV. Rather than
members (Mode IV).However, three possiblecondi- includethree equations for each of the latter two
tions may exist within both Modes I11 and IV depend- modesin the Specification, it wasassumed that Mode
ing upon whether the maximum bending occurs in the I11 consistedonlyofyieldingofthe threaded portion
shank or threaded portion of the screw or bending ofthescrewinthemainmember and that Mode IV
occurs at a location other than at the point of maxi- consisted only of yielding of the shank portion of the
mum moment (1 14,117). screw in the side member and of the threaded portion
Behavioral equations for each of the seven possible of thescrewin the mainmember(117). It was found
conditions were developed (1 14,117)and used to predict that thissimplificationcould be accomplished, while
the joint designvaluesofall configurations testedin obtaining approximately the same overall average ratio
the
original
research
(134).
Values of percent
5 ofyield mode design value to previous design value as
diameter offsetdowelbearing strength (Fe) required that resulting from use of three equations for each
for these equations were estimated from thespecific Mode, by assuminga constant ratio ofyield moment
gravityvalues of the test material using equations of the threaded portion to yield moment of the shank
approximately equivalent to thoseused for bolts (see of0.75(114,117). The yield moment ratio (Rm) for
Commentary for 8.2.1 - Yield Mode Equations). Yield thescrew appears intheselectedbehavioral equations
modedesignvalues for the test configurations were for Modes 111, and IV (1 14,117) as shown below.
compared to those obtained from the original method- Mode I
ology (equations C9.3-1 and C9.3-2) to verify there was
a reasonably stable relationship between the two for the P = D tsF,, (C9.3-3)
range of joint configurations available (1 14).
Mode 111,
Reexamination of the originallagscrewresearch
(134)showed that although the F 3 factor (the adjust-
ment for length of lagscrewin the mainmember)in
the original lag screw equations was related to ultimate
load or strength of the joint, the proportional limit
load (approximately one-fourth the ultimate) was

118 Lagsmws

2(1+R,)
2RmFyb(2+R,)D2
&=-I+ +
3Fcm t,2
Mode IV
where:
t, = thicknessofsidemember
D = unthreaded shank diameter of lagscrew
Fern= dowelbearing strength ofmainmember
F,, = dowelbearing strength ofsidemember
Rm = ratio ofyieldmomentof threaded portion
to yield moment of shank
Yield Mode Design Equations
Substituting 0.75 for Rm in the foregoing behavior-
al equations for Modes IIIs and IV and adding the
conversion factors relating yield mode design values to
previouslypublisheddesignvaluesgives the lagscrew
parallel to grain lateral designvalue ( 2 ) equations
givenin9.3.2of the Specification. The factors 4Ke ,
2.8Ke and 3Ke in the denominators of Equations 9.2-
1, 9.2-2 and 9.2-3, respectively, are the adjustments that
convertyield mode designvaluesbased on 5percent
diameter offsetdowelbearing strength to the levelof
proportional limit based design values tabulated in the
1986 and earlier editions of the specification. For
parallel to grain loading, Ke equals one. For perpen-
dicular to grain loading, Ke equals 1.25 for a connec-
tion with one member loaded parallel to grain and one
member loaded perpendicular to grain.
Dowel bearing strength values (F, ) used in the lag
screwyield mode equations are givenin Table 9A for
all structurally graded lumber species. The valuesalso
apply to glued laminated timber. The valuesin Table
9A have been established using the same dowel bearing
strength equations used for bolts (see Commentary for
8.2.1 - Yield Mode Equations). For perpendicular to
grain loading, the dowel strength equation isentered
with the shank diameter of the lag screw.
Abendingyield strength, F , of45,000psimay
be used for common lag screws Kbaving a shank diame-
ter of3/8inch or larger. For smaller diameter screws,
use ofhigheryield strengths maybe appropriate (see
Appendix I and Tables 9.3A and 9.3B). The yield
mode equations are applicable only to connectionsin
NDS Commentmy
which the total penetration of the lag screw (shank and
threaded portion) in the main member is at least four
times the shank diameter (4D), excluding the length of
the tapered tip, and the minimum thread length is that
specified for the diameter in Appendix L. This Appen-
dixprovides standard shank and root diameters and
total, thread and tapered-tip lengths for different screw
sizes. Tapered-tip lengths are calculated as 0.866the
root diameter.
It is to noted that when the length of the threaded
portion ofthe lag screwis greater than the standard
threaded portion length giveninAppendixLofthe
Specification, the root diameter rather than the shank
diameter should be used as D in the yield mode
equations if the boundary between the shank and
threaded portion of thescrewfallswithinthecleat or
side member.
For each joint configuration, thenominaldesign
value, 2,for each yield mode is calculated to determine
the limitingvalue for the connection. In mostcases,
Mode 111, or IV will be the limitingcase. For a
memberloaded at an angle to grain, the lagscrew
yield mode equations are entered with a dowel bearing
strength, Fee, calculated in accordance with Equation
9.3-4,a form of the standard bearingangle to grain
formula (see Appendix J). Design values for lag screws
acting at an angle to grain havebeenbasedonthis
equation using allowable lag screw parallel and perpen-
dicular to graindesignvalues as thereferencedesign
value levels since the 1944 edition (see Commentary for
8.2.1 - Member Loaded at an Angle to Grain).
TabulatedWood-to-WoodDesign Values. Lag
screwdesignvaluesgivenin Table 9.3Aassumethe
threaded portion of the screwislocatedcompletelyin
the main member, the total penetration of the screw in
the main member is a minimum of 8 0 (excluding the
length of the tapered tip), and screwsof standard
diameters, threaded lengths and other dimensions (see
AppendixL) are used. Valuesalso are based on lag
screwbendingyield strength valuesof45,000psi for
screwsof3/8inch and larger diameter, 60,000psi for
5/16inch diameter and 70,000 psi for 1/4inchdiame-
ter. The valuesassumed for the screwsizesless than
3/8inch diameter are thoseestimated from tests of
common wire nails of the same diameter (see Appendix
I). It is the responsibilityof the designer to assure
that thelag
screws
specified and used qualify for
tabulated designvalues.
Two lateral design values for perpendicular to grain
loading are shownin Table 9.3A: one for lag screw
connections with the side member loaded perpendicular
to grain and the main member loaded parallel to grain
A?DS Commentary

(Z,,); andonefor aconnectionwithmainmember Table C9.3-1- Comparison of 1991 and 1986 NDS
loadedperpendicular to grain and thesidemember Wood-to-Wood Single ShearLag Screw Lateral
loaded parallel to grain (Zm, ). Design Values
Comparison of 1991 and
Earlier Edition
Pent
Values. Lagscrewdesignvaluesgiveninthe1991
edition cover a greater range of side membertration geometries Side
thanthosegiveninearliereditions.Becauseofthe Member Lag Screw Depth Lag Screw Lateral DesignValue, lbs
Thickness Factor
simplifications made inthe yield mode equations and in. L D cd zll
the averaging procedures used to adjust yield mode lag in. in. (1991) 1991 1986 Ratio. 1991 1986 Ratio
screw design values to design values previously tabulat-
Southern Dinq:
ed, newdesignvalues are both higher and lowerthan
thosegiven for thesame joint configurations in the 1-ln 4 114
1.OOO
230
170 1.35 180 170 1.06
earliereditions. This isshown by thecomparisons in 318 0.760 3 19 250 1.28 228 190 1.20
TableC9.3-1. The penetration depth factor, c d , 1/2 0.516 366 290 1.26 217 190 1.14
showninthistable,applicable to the 1991 tabulated 6 114 1.OOO 230 230 1.00 180 220 0.82
values, is the adjustment for less than full design value 318 1.OOO 420 420 1.00 300 320 0.94
penetration ( 8 0 ) of the screw in the main member (see 112 1.OOO 710 600 1.18 500 390 1.28
9.3.3
of
Specification and foregoing Commentary). 2-112 7 112 1.ooO 750 650 1.15 520 420 1.24
Washerthicknessof1/8inch for 1/2inch and larger 3140.64610858501.28 704 470 1.50
diametersscrews and 0 inch for smallerdiameters 9 112 1.OOO 750 830 0.90 520 540 0.96
screwswereassumedincalculatingvaluesof cd . 314 0.979 1645 1350 1.22 1067 740 1.44
Adjustments for penetration depth are embedded in the 1 0.711 1820 1660 1.10 1180 830 1.42
values tabulated inpreviouseditions (see Equations SDruce-Pine-Fir:
C9.3-1 and C9.3-2).
1-ln 4 114
1.OOO
200
130
1.54 150 130 1.15
In termsof the designvalue ratios for the joint 318 0.760 281 180 1.56 198 140 1.41
configurations compared, lag screwdesignvalues for 112 0.516 294 210 1.40 206 140 1.47
wood-to-wood connections
based on 1991 edition 6 114 1.OOO 200 210 0.95 150 200 0.75
provisions average 22 percent and 25 percent higher for 318 1.OOO 370 370 1.00 260 280 0.93
parallel to grain loading and perpendicular to grain ln 1.OOO 570 430 1.33 400 280 1.43
loading,respectively, than those tabulated inthe1986 2-112 7 112 1.OOO 650 460
1.41 430 300 1.43
edition. The overallhigherdesignvalues for parallel 3140.6468796101.44 575 330 1.74
to grain loading (2 ) are aresultoftheprocedure 9 112 1.OOO 650 740 0.88 430 480 0.90
used to translate yie!h limitdesignvalues to the level 314 0.979 1331 960 1.39 871 530 1.64
of proportional limitbaseddesignvaluestabulatedin 1 0.711 190 1190 1.26 939 590 1.59
previous editions and to the equations used to establish
dowel bearing strength values. This procedure included
considering ratios ofyieldlimit and previousdesign member bearing strength, member thicknesses, fastener
diameter and fastener strength.
values forboth joints withwood and withsteelside
members when establishing a uniform conversion factor For joints made with two different species, tabulat-
for eachyield mode equation (see Commentary for eddesignvalues for thespecieswiththelowerdowel
9.3.1 - Yield Mode Equations). In thecaseofperpen- bearing strength may be used in lieu of using the yield
dicular to grain loading (2' ), the overallhigherlevel mode equations of9.3.1withthe appropriate dowel
ofdesignvaluesinthe1991editionreflectstheeffect bearing strength for eachspecies.
ofthe equation used to establishthedowelbearing
strengthvalues for perpendicular to grain loading in 9.3.2-Wood-to-Metal Connections
Table 9A. It is to be noted that lag screw perpendicu-
9.3.2.1 In previouseditions,designvalues for lag
lar to graindesignvalues tabulated inearliereditions
screw joints made withmetalsideplates were estab-
of the Specification were
based on application of
lished as 1.25 times the value for a wood-to-wood joint
procedures originally developed for bolts rather than on
ofequivalentconfiguration as determinedfromEqua-
I

tests of lagscrew joints under perpendicular to grain

tionC9.3-1 or C9.3-2(134).Metalsideplateswereas-
loading. The yield mode equations provide fully
a
sumed to be1/2inchthick and resultant valueswere
rationalizedbasis for evaluatingtheinteractions of
considered applicable to side plates of lesser thickness.

120 Lag Scmws

 NDS Commentary

For thickersideplates, tabulated designvalueswere Table C9.3-2 - Comparison of 1991 and 1986 NDS
required to be reduced for the lesser penetration of the Wood-to-Metal SingleShear Lag Screw Lateral
lagscrew. Design Values
Under the methodology of the 1991 edition, lag
screwdesignvalues for joints made withmetalside Stccl Pene
tration Side
plates are determinedusingthe Mode IIIs and Mode
IV equations for wood-to-wood joints in 9.3.1 using a Member Lag Screw Depth Lag Screw Lateral Design Value, Ibs
Thickness Factor
dowelbearing strength, Fcs , applicable to themetal in. L D cd zil Zrnl
used in the side plates (see Appendix I) and the actual in.
in. (1991) 1991 1986 Ratio 19911986 Ratio
thicknessoftheplate as thesidememberthickness.
As previously noted (see Commentary for 9.3.1 -
Background),the factors used to relate yield mode 1/4 4 1/4
1.OOO
310
240
1.29 230 180 1.28
design values to proportional limit based design values 318
1.OOO
510
4801.06 350 290 1.21
tabulated in previous editions were based on lag screw 1/20.8596976101.14 447 320 1.40
joints madewith both wood and metalsideplates. 6 3/8 1O
.OO 510 550 0.93 350 330 1.06
Thus Equations 9.3-2 and 9.3-3apply to joints with 1/2 1.OOO 810 950 0.85 520 490 1.06
metal as well as wood side plate when the appropriate 314 0.875 1452 ls00 0.97 840 660 1.27
input variables are used. 8 1/2 1.OOO 810 980 0.83 5205101.02
3/4 1.OOO 1660 2130 0.78 9609401.02
The lag screwdesignvalues for joints madewith 7/8 1.OOO 2220 2480 0.90 1240 1030 1.20
steelsideplatesgivenin Table 9.3Bassumethesame
strength lag screws as thoseused for Table 9.3A,a 10
5/8 1.OOO 1190 1550 0.77 7207400.97
dowelbearing strength of58,000psi for 1/4inchsteel 3/4 1O
. OO 1660 2200 0.75 9609700.99
1 1.OOO 2870 3680 0.78 1540 1470 1.05
(ASTM A36) side plates, and a dowel bearing strength
- of
45,000psi for steelplateslessthan1/4inch. The Spruce-Pine-Fir:
latter valueis the tensile strength for ASTMA446, 114 4 1/4 1.OOO 280
210 1.33 190 160 1.19
Grade A galvanized sheet steel. It is to be noted from 3/8 1.OOO 4503401.32 290 210 1.38
Table 9.3B that, for constant screw diameter and steel 1/20.859619 440 1.41 369 230 1.60
strength, designvaluesbased on the yield mode equa- 6 3/8 1.OOO 450 490 0.92 290 300 0.97
tions
decrease as steelplate
thickness decreases. 1/2 1.OOO 720 760 0.95 430 400 1.08
Previousmethodologyrecognized no effectof plate 314 0.875 1295 1070 1.21 709 470 1.51
thicknessbelow1/2inch. 8 1/2 1.OOO 720 880 0.82 430 460 0.93
Comparison of 1991 and Earlier Edition 314 1.OOO 1480 1560 0.95 810 690 1.17
7/8 1.OOO 1980 1770 1.12 1030 740 1.39
Values. Differencesin 1991 and earliereditiondesign
values for wood-to-metal lag screw joints are illustrated 10
5/8 1.OOO 1060 1380 0.77 6106600.92
in Table C9.3-2. The 1991designvalues are for joints 314 1.OOO 1480 1970 0.75 8108700.93
madewith1/4inchsideplates. The 1986values apply 1 1.OOO 2550 2640 0.97 1300 1050 1.24
to joints made with1/2inch or thinnersideplates.
The valuesof Cd less than 1.000in Table C9.3-2 are Commentary for 9.3.1 - Yield Mode Equations, Com-
the adjustments applied to 1991 tabulated design values parison).
when thereisless than fulldesignvaluepenetration
( 8 0 ) of thescrewin the mainmember(see9.3.3 of 9.3.2.2 (See Commentary for 7.2.3)
Specification and foregoingCommentary). Joints were 9.3.3-Penetration DepthFactor, Cd
assumed made without washers.
The penetration depth factor provides for reduced
The averagedesignvalue ratios for theconfigura- design values when the length of penetration of the lag
tionscomparedshow 1991designvalues for metal-to- screwinthemainmember, bothshankand threaded
wood connections were 2 percent lower and 16 percent portion, isless than eighttimesthe shank diameter
higher for the parallel and perpendicular to grain ( 8 0 ) . The full proportional limitdesignvalue of lag
.- loading
cases,
respectively,
than
the
comparable 1986 screw joints of anyspeciesisconsidereddeveloped
designvalues. The higheroveralldesignvalues for the whenthispenetration depth occurs (seeCommentary
perpendicular to grain comparisons reflect the effects of for 9.3.1 - Background, 1991 Edition).Penetrations
the equation used to establishdowelbearingstrengths
down to 50 percent or 4 0 ofthefulldesignvalue
for perpendicular to grain loads in Table 9A (see

Lag Screws 121

NDS Commentary

penetration depth are allowed if tabulated design values and 4.167 are the adjustments used to convert the two
or yield mode equation design values are reduced types of test values to allowable levels (see Commentary
proportionately. for 9.2.1 and 9.3.1 - Background).
The penetration depth of the lag screw in the main To account for the interaction observed in the new
member is calculated as the length of the lag screw (P) lag screwtests, Equation 9.3-6 has been introduced in
minus the sum of the thicknessof the sidemember the 1991 edition for determining the allowabledesign
(t- ), the thicknessof any washer used (w) and the value of lag screwssubject to combined lateral and
lengthof the tapered tip of the screw ( E ) , or withdrawal loads. This equation, a form of the bearing
[P- (t- + w+E)J. Lag screw dimensions, including thread angle to grain equation (see Appendix J), is
and tapered tip lengths, are giveninAppendixL.
Consistentwith the basisoflagscrewdesignvalues 2; = ( W'P)Z' (C9.3-6)
publishedinprevious editions, awasherthicknessof ( W' p ) cos2 a +z'sin2a
1/8 inch for screw diameters of 112 inch and larger and
0 inch for smaller diameter screwsmaybeassumed. where:
Examplesof the use of the Cd factor to adjust Za' = allowabledesignvalue for lagscrewloaded
designvalues for less than full penetration are shown at angle to the surface of mainmember
in Tables C9.3-1 and C9.3-2 and related Commentary.
2' = lateral designvalue for lagscrewconnec-
9.3.4-End Grain Factor, CCg tion
Use of two-thirds the perpendicular to grain lag W' = withdrawal
design
value for lagscrew
screw lateral design value for screws inserted in the end connection per inch of penetration
grain of the main member (62) has been a provision of
theSpecificationsince the 1944 edition. Whendesign p = length of thread penetration in
the
main
values for this case are based on the yield mode equa- member
tions of 9.3. l, the dowel bearing strength of the main a = anglebetweenwoodsurface and direction
member is assumed equal to the perpendicular to grain
of applied load
bearing strength of the species from Table 9A.
The length of penetration ofthe threaded portion
Because of the tendencyof the member to split of the screwin the mainmemberexcludesthelength
under lateral loading, structural lagscrewconnections
of the taperedtip and includes the reductioninpene-
inend grain surfaces should beavoided(62,66).Lag
tration resulting from the use of awasherunderthe
screws in end grain surfaces particularly should not be
screw head.
subjected to combined withdrawal and lateral loading.
Equation C9.3-6 will give
generally conservative
9.3.5-Combined Lateral and Withdrawal Loads
designvalues for load angles greater than 45". Equa-
In the 1977 through the 1986 editions, the Specifi- tion C9.3-6 can also be used to determine the allowable
cation provided that lagscrewssubjected to combined designvalueoflagscrewsembedded at an angleto
lateral and withdrawal loads be analyzed separately for grain inthewoodmember and loaded inadirection
theresistance of the screw to each load. The results normal to the woodmember. For thiscondition a
ofrecent lag screw tests (115) showed that withdrawal would be defined as the angle between
thewood
load components did not reduce lateral load capacity surface and thelagscrew as shownin Figure C9.3-1.
when maximum joint loads are considered.However,
when joint resistance was evaluated at the design load
levelby expressing the strength of the joint loaded at
any angle to the surface as the lesser of the proportion-
al limit load divided by 1.875(2.25/1.2) or themaxi-
mum load divided by 4.167 (5/1.2), an interaction of
the load components was observed with larger diameter
screws at load angles less than 45" (1 15). Use of these
alternative design load basesisrequiredbecauselag
screw lateral designvalues are basedon proportional
limit joint loads whereas withdrawal design values are Fzgure 0 . 3 - 1 Combined Lateraland Withdra walLoadforLag
based on maximum joint loads. The factors of 1.875 Screwlnsertedat an Angle to Wood Surfcz

122 Lag h w s
 MJS Commentary

9.4-PLACEMENT OF LAG SCREWS was

9.4.1-GeometryFactor, Edge Distance,
End Distance, Spacing
Applicationof the samepositioningrequirements
for lagscrews as those for bolts has beenaprovision
of the Specification since
the 1944 edition. The
similarity of the performance characteristics of the two
typesoffastenerswasrecognizedintheoriginallag
screw research (134) and the use of edge distance, end
distance and spacing criteria for bolts withlagscrews
specifically
recommended (62). Lag screw tests
conducted on Douglas-fir joints in 1963 confirmed that
a 4 0 spacingbetweenlagscrewsinarowwasmore
than sufficient to developthefull proportional limit
load capacity of the joint (102).
9.4.2-Multiple Lag Screws
The group action factor, Cg, has beenappliedto
lagscrew joints containing two or morescrewsina
row(screwsalignedinthedirectionofthe load) since
the factor wasfirstintroducedin 1973.