Beruflich Dokumente
Kultur Dokumente
MEMBER
FELLOW
ASAE
ASAE
ASAE
ABSTRACT
Metal-plate-connected wood trusses with semi-rigid
joints were investigated by the matrix method of structural
analysis. The element-stiffness matrix and fixed-end forces
of an individual member with one or both ends semi-rigid
were derived as modifications of the idealized cases. The
case of unequal elastic connections at two ends of a
member can also be handled by this method of analysis.
When one end is semi-rigid, the other end may be specified
as pinned, rigid, or semi-rigid. The truss was analyzed for
three different joint assumptions: pin, rigid, and semi-rigid
joints. The truss performance, based on deflection, varied
greatly depending on the joint assumption. Including semirigid-joint behavior in the analysis of a wood truss
decreased maximum deflection by 34% compared with the
pinned-joint assumption and maximum moment by 13%
compared with the rigid-joint assumption. By incorporating
semi-rigid behavior of joints, more accurate member forces
can be obtained enabling closer prediction of actual truss
behavior. KEYWORDS. Wood engineering. Matrix method.
Element-stiffness matrix. Fixed-end forces.
INTRODUCTION
1011
3
5
7
BOTTOM CHORD DEAD LOAD 292 N/m (10 psf)
Figure l~Fink truss analog with design loads. Joints 1 and 8 are heel
Joints; 3 and 7 are web joints; and 5 is a tension splice joint
MODELING JOINTS
the other (j-end). Let SR be the rotational stiffness of the iend. Rotational stiffness is defined as the moment per unit
relative rotation (N-m/rad) at the connection. The length of
the member is L, moment of inertia I, and E is the modulus
of elasticity (MOE). The compatibility approach (Weaver
and Gere, 1986) was used to derive FEF: Ri, R2, M^, and
M2 (see fig. 2a).
For the beam in figure 2a, the j-end is assumed free so
that the reactions R2 and M2 are zero. This statically
determinate beam is shown in figure 2b. Transverse
deflection and rotation at the j-end are determined as
follows.
Transverse deflection at the j-end, dj, is composed of
the following two parts:
1. Deflection due to load q = - q L^/S E I
(1)
(2)
(3)
(4)
MODIFICATION OF FIXED-END-FORCES
Y-axis
q N/m
^
i-end
X-axis
*
*":
j-end
(5)
(7)
(8)
Mo
ih^
(a)
(9)
(10)
(11)
(12)
Figure 2~Beam witli one end semi-rigid and tlie otiier end rigid:
determination of the fixed-end-forces.
1012
Rl+Mi=qL
(13)
(23)
(14)
(15)
(16)
= M2L2/2
EI
(17)
Mi=qL2/12e4
Therefore,
Rj = q Le3/2 e^
R2 = q L e5/2 c^
Mi=qL2/12e4
M2 = - q L 2 e6/12e4
{ok}
(18)
Rl=3qL/8
Mi=0
R2 = 5qL/8
M2=-qL2/8
(19)
(ok)
1013
Y-axis
E, I.A, L
Z-axis y
i-end
Ri
j-end
(a)
Ai
(34)
(35)
X-axis
Di2 = (L2/2EI) +
Therefore,
(36)
(L/SRJ)
The rotation of the j-end due to a unit moment at the j end is denoted by D22 and is composed of the following
three parts:
1. Rotation due to the unit moment = L/EI
2. Rotation due to the rotation at the i-end =
(37)
I/SRJ
(38)
(39)
ejy = EI/LSRJ;
(40)
^Ril - 1 + Ri'
^Ri2 = 1 + 2 ejy;
eRj2=l+2eRj;etc...
(26)
(27)
DH=L3eRi3/3EI;
Di2 = L2eRi2/2EI;
I/SRJXL
(28)
Dj j = L3/3EI + L2/SRi
(29)
(41)
Therefore,
[Fi] =
= JL.
6EI
Therefore,
(31)
= L2/2EI
(32)
1014
(33)
D22
2L eRj3 3LeRi2
(42)
3Lej^i2
6ejm J
3LeRi2
[s.jc
(43)
- 3LeRi2 2L e^ij
Where, C =
D21
2EI
L'[4eRij + 3 ( 4 e R i e R j - l ) ]
(44)
[S'M] =
where: [Sy], [Sy], [Sjj], and [Sjj] are all 2x2 submatrices of
[S'M]. The terms in [Sjj] are defined as the restraint actions
at the j-end of the member due to unit displacements at the
same end. The terms in [Sy] are restraint actions at the
i-end due to unit displacements at the j-end, and they are in
equilibrium with the terms in [Sj,]. The terms in [Sjj]
consist of restraint actions at tne j-end due to unit
displacement at the i-end. The terms in [Sjj] are restraint
actions at the i-end due to unit displacements at the i-end,
and they are in equilibrium with the terms in [Sjj.
The matrix [Sjj] has already been found and given by
equation (43). The other three submatrices of [S'j^] can be
found b.y the transformation of axis. Statically equivalent
forces at the i-end may be computed using the
transformation matrix, [T], where [T] is:
[T]=[-]
(45)
eAi = EA/LSAi;
3LeRi2
[s] = c
(46)
-3LeRj2
(50)
L^
eAj = EA/LSAj;
(51)
[Kj] = [d]-l=AE/LeAij
(52)
(53)
[T] = [1]
-3LeRj2
[s,] = c
(47)
(54)
[s]=c
6ejm
3Le,y2
(48)
^^j2
2L ej^j3 J
^Wj2
""^Rij
3Lej^i2
AE
LeAij
-1
-1
(55)
Y-axis
A.E.L
j-end,
X-axis
(a)
[S'M] = C
3LeRj2
2 L ej^j3 - 3 L ^ j 2
(49)
^
^7^
(b)
777^
L'
"3Lei,i2 2L2eRi3
1015
AE
KJ
L^Aij
[SM]
AE
LCAU
6CeRij
3CLej^j2
-eccRij
3CLejji2
3CLeRj2
2CL2eRj3
-3CLeRj2
CL2
(56)
=
AE
AE
LeAij
LCAij
-6CeRy
- 3Q^Rj2
6CeRij
-3CURi2
3CLej^j2
CL2
-SCLe^jj
2CL2eRi3
Element stiffness matrices for other end conditions are ANALYSIS METHODS
determined from equation 56 by taking appropriate limits.
The truss shown in figure 1 was analyzed for the
For example, for both ends rigid, Sjy = SR: = > and S^^j = following three different joint models:
Sj^i = oo. Then C becomes lElfL^ and Cj^j = c^: = 0; e^^^j =
Truss type analysis (PP). Web members are pinned at
both ends to continuous top and bottom chords. The
Aj "^ 0 ^ i i ~ Ril = ^Rjl ~ ^' ^^^ Aij" ^' Substituting the
above conditions into equation 56 yields:
[SM]
AE
L
_AE
L
12EI
6EI
L2
_12EI
6EI
L2
6EI
L2
4EI
L
6EI
L2
2EI
L
AE
L
12EI
_6EI
L2
=
AE
L
0
12EI
_6EI
L2
6EI
L2
2EI
L
A fink truss, which is the most commonly used metalplate-connected wood truss in light frame residential
construction, was analyzed in this study (see fig. 1). The
truss has a 5:12 slope, 8.53-m (28-ft) span (L), and is
spaced 0.61 m (2 ft) on center. All tmss members were 38
X 89 mm ( 2 x 4 in.) SP No. 2 KD 15 and were connected
by 20-gage punched metal plates. Two 76 x 127-mm ( 3 x 5
in.) plates were used for each heel joint and web joint at the
bottom chord and two 76 x 102-mm ( 3 x 4 in.) plates were
used for the tension splice joints. The loads used to analyze
the truss are summarized in figure 1. Load duration factor
of 1.15 was used in this study.
1016
em
L2
(57)
4EI
L
Rotational Stiffness
Joint
(lb/in.)
(lb-in./rad)
Heel
15.0 x l O ^
38.0 x l O ^
Tension
3.0 x l O ^
14 x l O ^
Web
14 xlO^
2.0 X 10^
TABLE 3. Forces at i- and j-ends and CSI in a top chord member (6-8)
and bottom chord member (1-3)
Member 6-8
PP
6.64
(0.2614)*
15.78
(0.6213)
0.01307
RR
Force
Axial (N)*
Shear(N)
SR
6.42
(0.2528)
8.35
(0.3287)
6.94
(0.2732)
10.39
(0.4091)
0.01115
0.01195
Moment (N-m)t
CSlt
Member 1-3
End
SR
RR
PP
RR
PP
i-end
11583
11677
11726
11178
11276
11339
j-end
-12482
12575
-12629
11178
11277
11339
i-end
1259
1299
1335
294
297
311
j-end
903
867
827
538
535
520
i-end
493
567
-82
-69
j-end
-82
-69
-270
-268
2%
1.28
1.42
0.84
0.85
0.89
1.45
SR
* IN
= 0.225Ibf.
t IN-m = 8.851 Ibf-in.
I For adequate design, the CSI values should be less than or equal to 1.0
(National Forest Products Association, 1986).
The analysis in this study is only meant for comparison.
1017
SUMMARY
REFERENCES
1018