Fundamentals of Vibration Theory
Fundamentals of Vibration Theory
& m * && + c * x + k * x = 0 x
where: x: m: c: k: displacement mass damping constant spring stiffness
(2.1)
Figure 21: Point mass with damping and spring under dynamic excitation The terms in Eq. (2.1) represent forces standing in equilibrium: m * && inertial force x
The external force F(t) in this specific case is zero, the movement initiates with an impact or its deformation. The differential equation can be solved with an exponential type of solution:
-5-
x(t ) = U * e s*t
& x(t ) = U * s * e s*t &&(t ) = U * s 2 * e s*t & x
X and its derivates are substituted in Eq. 2.1:
m * s2 + c * s + k = 0
The undamped circular Eigenfrequency and the damping ratio are introduced:
(2.5)
k m c = 2m1
12 =
(2.6) (2.7)
The damping ratio relates the existing damping to the critical damping
ccrit = 2 * m * 1
(2.8)
A damping is called critical when the amplitude decays from the maximum displacement to the rest position without changing the sign without oscillation:
1,5 1,5
0,5
-0,5
-0,5
-1
-1
-1,5 t
-1,5 t
-6-
In civil engineering the damping ratios are low, with <<1 (usually about 0,05 - 0,10). Thus, when solving Eq. 2.5, the roots are complex:
2 s = * 1 ( * 1 ) 2 1 = 1 * ( 2 1)
(2.9) (2.10)
s = 1 * ( i 1 2 )
When substituting Eq. (2.10) into (2.2), the displacement solution follows as:
(2.11) (2.12)
UB =
XB +i* XA 2
1 = arctan
XA = phase angle XB
X A = 0, X B =1
Eq. (2.18) reduces to
(2.19)
& m * && + c * x + k * x = F (t ) x
F (t ) = F0 * sen( 0 * t )
(2.20) (2.21)
-7-
XC =
XD =
F0 1 ( 0 / 1 ) 2 * k 1 ( / ) 2 2 + (2 * * / ) 2 0 1 0 1
(2.24)
F0 2 * * 0 / 1 * k 1 ( / ) 2 2 + (2 * * / ) 2 0 1 0 1
(2.25) (2.26)
x(t ) = X 0 * sen( 0 * t + 0 )
2 2 X0 = XC + X D
0 = arctan
XD XC
(2.27) and
(2.28) (2.29)
Figure 26: MDOF system This simplification is invalid in the case of vertical vibration within one floor. Now the mass and stiffness distribution play an important role, and the differential equation is written in matrix-form:
~ ~ ~ ~ & * && + C * x + K * x = F x
where M mass matrix K stiffness matrix C viscous damping matrix x displacement vector F forcing function vector
(2.30)
In general purpose with distributed masses and stiffnesses, this equation is to solve by means of finite element methods.
-8-
(EI * w( x)) = q( x)
with EI = constant and q = constant simplifies to
EI * w( x) IV = q
integration yields
EI * w( x) = q * x + c1 = Q ( x)
knowing that Q ( x = 0) =
q *l 2
(3.4), gives c1 =
q *l (3.5), so that 2
(3.6)
EI * w( x) = q * x
q *l 2
integration yields
EI * w( x) =
(3.7)
(3.8) (3.10)
EI * w( x) =
1 q *l 2 q * x3 * x + c2 * x + c3 6 4
(3.12) or w( x = 0) =
(3.11)
so EI * w( x = 0) = c3 integration yields
c3 EI
(3.13)
EI * w( x) =
(3.14) (3.16)
As we assume two identical springs at the ends of the beam, the middle point has to be a maximum of deflection:
w( x = l / 2) = 0 =
1 l q *l l l q * * + c2 * + c3 6 2 4 2 2
(3.17)
0=
(3.19) (3.20)
c3 =
ql 3 c3 l k* * (3.21) 24 EI 2
kl ql 3 c3 * 1 + = 2 EI 24
c3 =
ql 3 * EI 24 EI + 12kl
(3.23)
c2 =
From (3.8) it is known that c2 = M ( x = 0) (3.25), so that M ( x = 0) = As the clamping moment is known as M Cl = the clamping ratio can be quantified as RM
ql 2 12 M = M Cl ql 2 ql 2 / 10 = 12 120
(3.29)
So when pretending a clamping ratio of 0,1 or 10 %, M equals to M = with M given, transforming (3.26) leads to: k =
24 EI * M ql 3 12M * l
With the goal to compare this theoretical relation, and to verify its application in the calculation program SAP2000, a comparison is documented in the following chapter.
In order to compare theoretical and numerical Eigen modes, a single-span beam with pinned ends was designed. Its rigidity E*I was defined as 1, the length is 1, and the distributed mass is 1. The clamping ratio is given in 20 steps of 5 %, from 0 equals to pinned ends until 1 which means full clamping. For each step it is calculated the spring rigidity as documented in the prior chapter. The constant mass of the beam causes a deformation and a clamping moment, which is calculated both analytically and numerically. Subsequently it is calculated the frequency of the first and the second Eigenmode. The results are as shown: Clamping (ratio of clamping moment)
Eigenfrecuency [Hz]
error deformation
0,00 0,05 0,10 0,15 0,20 0,25 0,30 0,35 0,40 0,45 0,50 0,55 0,60 0,65 0,70 0,75 0,80 0,85 0,90 0,95 1,00
0,0 0,105 0,222 0,353 0,500 0,667 0,857 1,077 1,333 1,636 2,000 2,444 3,000 3,714 4,667 6,000 8,000 11,333 18,000 38,000
0 0,004 0,008 0,013 0,017 0,021 0,025 0,029 0,033 0,038 0,042 0,046 0,050 0,054 0,058 0,063 0,067 0,071 0,075 0,079 0,083
0,0E+00 8,0E-10 4,0E-10 1,4E-16 2,0E-09 1,6E-09 1,4E-16 1,1E-09 1,0E-09 0,0E+00 8,0E-10 7,3E-10 1,4E-16 6,2E-10 5,7E-10 2,2E-16 5,0E-10 4,7E-10 3,7E-16 4,2E-10 4,0E-10
0 0,004 0,008 0,013 0,017 0,021 0,025 0,029 0,033 0,038 0,042 0,046 0,050 0,054 0,058 0,063 0,067 0,071 0,075 0,079 0,083
130,21 125,00 119,79 114,58 109,38 104,17 98,96 93,75 88,54 83,33 78,13 72,92 67,71 62,50 57,29 52,08 46,88 41,67 36,46 31,25 26,04
2,5E-04 2,6E-04 2,7E-04 2,8E-04 3,0E-04 3,1E-04 3,3E-04 3,5E-04 3,7E-04 3,9E-04 4,2E-04 4,5E-04 4,8E-04 5,2E-04 5,7E-04 6,2E-04 6,9E-04 7,8E-04 8,9E-04 1,0E-03 1,2E-03
130,24 125,03 119,82 114,62 109,41 104,20 98,99 93,78 88,57 83,37 78,16 72,95 67,74 62,53 57,32 52,12 46,91 41,70 36,49 31,28 26,07
8,76 8,94 9,14 9,34 9,56 9,80 10,05 10,33 10,63 10,95 11,31 11,71 12,15 12,65 13,21 13,86 14,61 15,49 16,56 17,89 19,60
0,64 0,62 0,61 0,60 0,58 0,57 0,55 0,54 0,52 0,51 0,49 0,47 0,46 0,44 0,42 0,40 0,38 0,36 0,33 0,31 0,28
0,16 0,16 0,16 0,16 0,16 0,15 0,15 0,15 0,15 0,15 0,15 0,14 0,14 0,14 0,14 0,13 0,13 0,12 0,12 0,11 0,10
4,00 3,94 3,88 3,81 3,75 3,68 3,62 3,55 3,49 3,42 3,35 3,28 3,21 3,14 3,07 3,00 2,93 2,87 2,82 2,77 2,75
Figure 27: Table Data relation springs - clamping ratio Analyzing the results a good accordance is found in terms of deformation and clamping moment:
- 11 -
Relation T2/T1
T1 [s]
T2 [s]
Figure 28: Spring rigidity as a function of clamping ratio The relation of clamping ratio and spring rigidity is no-lineal, going from a spring rigidity zero until infinite. This is why the first and the last step are not shown in the diagram above.
Figure 29: Relative error in clamping moment Concerning the clamping moment the error is basically caused by rounding. Whenever the value has less decimals than the maximum of decimals shown, the error turns out zero.
Figure 210: Relative error in centre deformation Concerning the centre deformation the error rises with the clamping ratio, still showing a good accordance between analytics and numerical approximation.
- 12 -
Figure 211: Relation second/first Eigenfrequency over clamping ratio Another conclusion of this parameter study is the relation between the clamping ratio and the relation of the first two Eigenfrequencies, going from 4 in the case of pinned ends to 2,75 in the case of clamped ends. So when analyzing measurement data of a single-span beam, one can determine the clamping ratio as a function of the relation between the first two Eigenfrequencies.
Figure 212: Comparison Eigen and Ritz-vectors The first 3 modes are determined with great accordance between both methods, for higher modes the differences increase notably. As this thesis refers to analysis and measurements, the exactness is satisfactory. In very little cases the measurement equipment and the general conditions permit to discover more than the two first modes.
- 13 -
Figure 213: Response spectrum effects on persons and structures [2] This graph is from Britisch Standard 6472, another standard used is german DIN 4150 part 3. Values are established as a function of building use and frequency:
Swiss norm SN 640312 fist categorizes structures and afterwards establishes these limits for each:
Figure 217: KB acceptation values [3] International standard ISO 2631-1 offers a simpler solution, giving maximum accelerations for every step from comfortable until extremely uncomfortable.
- 15 -