Design To Prevent Floor Vibrations-Thomas M. Murray
Design To Prevent Floor Vibrations-Thomas M. Murray
Design To Prevent Floor Vibrations-Thomas M. Murray
THOMAS M. MURRAY
HUMAN PERCEPTIBILITY
Human perceptibility to transient floor vibrations depends on
three factors: frequency, initial amplitude, and damping.
Although a number of scales have been developed to measure
human sensitivity to steady state (zero damping) vibration,2
only two scales are available which include the effects of
damping: the modified Reiher-Meister scale and the WissParmelee scale.
The modified Reiher-Meister scale, Fig. 1a, relates the
effects of amplitude and frequency to four levels of human
perception. The original Reiher-Meister scale was developed
for steady state vibrations.3 Lenzen,4 after studying 46 steel
joist-concrete slab floor systems subjected to single heel-drop
impacts, modified the scale by multiplying the amplitude axis
by a factor of 10 to account for the transient nature of the
vibration. For transient vibrations "amplitude" is defined as
the first maximum amplitude. Lenzen, therefore, implicitly
accounts for damping found in typical floor systems.
Although the Reiher-Meister curves were developed for
steady state vibration using a limited number of subjects and
have been shown to be statistically inaccurate,5 the modified
curves have been verified by different researchers for several
types of floor systems: Lenzen,4 steel joist-concrete slab
systems; Lenzen and Murray6 and Rahman and Murray,7
steel beam-concrete slab systems; Polensek,8 wood joist
systems; Commonwealth Experimental Building Station,9
various systems.
Wiss and Parmelee10 have conducted the only known
laboratory study to determine human perception to transient
vibration of the type normally found in floor systems. A total
of 40 humans were subjected to a waveform "designed to
simulate vibrations caused in floor systems by one foot fall."
Frequency, amplitude and damping were varied through a
range found in normal floor systems, and the following
human response formula was developed from statistical
analysis of the subjective ratings of the vibrations:
82
ENGINEERING JOURNAL/AMERICAN INSTITUTE OF STEEL CONSTRUCTION
2003 by American Institute of Steel Construction, Inc. All rights reserved. This publication or any part thereof must not be reproduced in any form without the written permission of the publisher.
fA
R = 5.08 00.217
D
0.265
(1)
where
R = the mean response rating
= 1, imperceptible vibration
= 2, barely perceptible vibration
= 3, distinctly perceptible vibration
= 4, strongly perceptible vibration
= 5, severe vibration
f = frequency
A0 = maximum amplitude
D = damping ratio
A comparison of the modified Reiher-Meister curves, taken
from Ref. 1, and the predictions from Eq. (1) is shown in Fig.
1b. The shaded regions were obtained by assuming R = 1.5,
2.5, and 3.5 by varying the damping from 4% to 10% of
critical (typical values for concrete slab-steel beam floor
systems). Since whole number R values represent the center
of perceptible regions, the shaded regions separate relevant
regions, as do the diagonal lines of the modified ReiherMeister plot. In all regions the Wiss-Parmelee human
response formula is more critical than the modified ReiherMeister curve, which can be explained from the fact that the
data was obtained in a laboratory from humans expecting
motion and knowing that they must judge it, while the
modified curve was developed from data taken from humans
subjected to actual, on site, floor motion. Because of this
difference, Ref. 5 recommends the use of the modified
Reiher-Meister plots for making judgments about vibration
perceptibility of steel joist concrete slab floor systems.
From measurements and subjective evaluations obtained
by the author, it has been found that steel beam-concrete slab
systems, with relatively open areas free of partitions and
damping between 4 and 10%, which plot above the upper
one-half of the distinctly perceptible range, will result in
complaints from the occupants and that systems that plot in
the strongly perceptible range will be unacceptable to both
occupants and owners. Although this conclusion cannot be
proven statistically, it substantially agrees with the modified
Reiher-Meister curvers and with the Wiss-Parmelee human
response formula, and is recommended for design use.
Frequency, CPS
(a) Modified Reiher-Meister scale
Frequency, CPS
(b) Comparison of modified Reiher-Meister and Wiss-Parmelee scales
Figure 1
FREQUENCY
(a) Composite action, regardless of the method of construction
(b) An effective slab width, S, equal to the sum of half the
distance to adjacent beams
(c) An effective slab depth, de, such that the rectangular slab used
to compute the moment of inertia is equal in weight to the
actual slab including concrete in the valleys of decking and the
weight of decking
83
THIRD QUARTER/1975
2003 by American Institute of Steel Construction, Inc. All rights reserved. This publication or any part thereof must not be reproduced in any form without the written permission of the publisher.
gEI t
f = 157
.
3
WL
(2)
where
g
E
It
W
L
= 386 in./sec
= modulus of elasticity, ksi
= transformed moment of inertia, in.4
= total weight supported by the tee-beam, kips
= tee-beam span, in.
The total weight W used in Eq. (2) should include all of the
dead load plus an estimate of the live load at times when
floor motion might be more annoying. For instance,
complaints have been received from grade school teachers
concerning motion of classroom floors after school hours
when most of the children had gone home. A reasonable
estimate for live load to be included in W is 10% to 25% of
the design live load.
AMPLITUDE
The impact of a 190-lb man executing a heel drop was
measured by Ohmart11 and is shown by the solid line in Fig.
3. An approximation to the actual impact is given by the
dashed line and has been shown to be sufficiently accurate.11
The amplitude of a single simply supported tee-beam
subjected to the approximate impact can be determined from
Eq. (3) or (4):
If t0 = (1/ f) tan1 0.05:
246 L3
Aot =
(010
. t0 )
EI t
(6)
Ao = Aot/Neff
(7)
(3)
If to > 0.05:
246 L3 1
EI t 2 f
For E
= 29 106 psi:
(4)
600 L3
48EI t
(5a)
= 29 103 ksi:
L3
(5b)
80EI t
where (DLF)max = maximum dynamic load factor, which can
be obtained from Fig. 4.
Aot = ( DLF ) max
84
ENGINEERING JOURNAL/AMERICAN INSTITUTE OF STEEL CONSTRUCTION
2003 by American Institute of Steel Construction, Inc. All rights reserved. This publication or any part thereof must not be reproduced in any form without the written permission of the publisher.
DAMPING
Although damping is the most important floor vibration
parameter, it is not presently possible to accurately predict
the damping that will exist in a floor system. Damping in a
floor system is thought to be influenced by the type thickness,
concrete weight, fire protection, partitions, ceiling, ductwork
of construction, slab, plumbing, etc. Very little research has
been conducted to determine the amount of damping
contributed by the
85
THIRD QUARTER/1975
2003 by American Institute of Steel Construction, Inc. All rights reserved. This publication or any part thereof must not be reproduced in any form without the written permission of the publisher.
DESIGN PROCEDURE
1. Estimate the damping in the finished floor system; if
greater than 810% there is no need for a vibration
analysis.
2. Compute the transformed moment of inertia of a single
tee-beam, It, using the model of Fig. 2.
3. Compute the frequency, f, from Eq. (2).
4. Compute the heel-drop amplitude of a single tee-beam,
Aot, using Eq. (5) and Fig. 4.
5. Estimate the effective number of tee-beams, Neff, using
Fig. 5.
6. Compute the amplitude of the floor system, Ao = Aot/Neff.
7. Plot on the Reiher-Meister scale, Fig. 1a.
8. Redesign if necessary.
Figure 6
Example
Single Tee-Beam Amplitude:
0.05f = 0.05 10.78 = 0.539
Given:
Check the following floor system for susceptibility to
vibration:
2-in. lightweight concrete slab (110 pcf, n = 14)
2-in. metal deck (concrete in deck + deck = 9.1 psf)
Span: 25 ft-6 in.; Beam Spacing: 8 ft-6 in.
W1430 A36 steel: A = 8.83 in.2, I = 290.0 in.4
Non-composite construction
Hung ceiling; very little ductwork
Solution:
Damping:
From Fig. 4:
(DLF)max = 1.22
Aot = ( DLF ) max
=
7.29(35
.)
= 0.0162 in.
+ 255
. (2.49)
12
+ 290 + 8.83[14.12 (13.86/2)]2
= 93 in.4
Ac/n = 102(3.5)/14 = 25.52 sq. in.
2
Redesign:
As shown in Fig. 7, increasing the beam size is not an
effective method for decreasing vibration perceptibility.
When the supporting beam is increased two depths, there is
very little change in the rating.
By increasing the slab thickness, a significant change is
obtained. As shown in Fig. 7, the floor system becomes
acceptable when the slab is increased to 3.5 in. If headroom
is available, a better design is a W1631 with 3.5-in. slab.
Frequency:
W = [(3.5/12) 110 8.5 25.5] + (30 25.5)
= 7719 lbs = 7.72 kips
386 29 103 931
gEI t
f = 157
.
=
157
.
WL3
. 12) 3
7700 (255
80 29 103 931
122
. (255
. 12) 3
L3
80EI t
= 10.78 cps
86
ENGINEERING JOURNAL/AMERICAN INSTITUTE OF STEEL CONSTRUCTION
2003 by American Institute of Steel Construction, Inc. All rights reserved. This publication or any part thereof must not be reproduced in any form without the written permission of the publisher.
Figure 7
REFERENCES
1. Galambos, Theodore V. Vibration of Steel Joist-Concrete Slab
Floors Technical Digest No. 5, Steel Joist Institute, Arlington,
Virginia.
2. Hanes, R. M. Transportation Programs Report, Human Sensitivity
to Whole-Body Vibration in Urban Transportation Systems: A
Literature Review APL/JHU-TPR 004, The Johns Hopkins
University, Applied Physics Laboratory, Silver Spring,
Maryland, May, 1970.
87
THIRD QUARTER/1975
2003 by American Institute of Steel Construction, Inc. All rights reserved. This publication or any part thereof must not be reproduced in any form without the written permission of the publisher.