The Steel Construction Institute

Vibration Characteristics of
Modern Composite Floor
Systems
Stephen Hicks and Paul Devine
Factors that have caused vibrations to be an
important consideration in modern design

Efficient long-spanning floors (comprising

composite cellular beams, etc.) tend to
possess low natural frequencies.
The advent of the ‘paper-free’ office has
reduced the permanent mass (from filing
cabinets, bookcases, etc.) that is present on
the floor.
Full-height partitions are becoming less
common, leading to reduced damping.
Directions for vibrations defined in
ISO2631
z

x
S u p p o rtin g
s u rfa c e
S u p p o rtin g
s u rfa c e

S u p p o rtin g
s u rfa c e y
 ISO 2631-1: 1997 & ISO2631-2:2003
frequency weighting factors for perception
10

Weighting factor

Wk (z-axis)

1 Wd (x- & y-axis)

Wm (z-, x- & y-axis

combined)

0.1
1 10 100
Frequency (Hz)
Multiplying factors specified by ISO/CD 10137
for ‘low probability of adverse comment’

Place Time Multiplying factor

Exposure to
continuous vibration
(16 h day, 8 h night)
Critical working areas (e.g., Day 1
hospital operating theatres, Night 1
etc.)
Residential Day 2 to 4
Night 1.4
General office Day 4
Night 4
Workshops Day 8
Night 8
Damping & Frequent Variable Action Factor

For design, it is recommended that the following damping

values may be assumed:

z = 1.1% for bare unfurnished floors.

z = 3.0% for floors in normal use.
z = 4.5% for a floor with partitions, where the designer is confident that
partitions will be appropriately located to interrupt the relevant
modes of vibration.

It is suggested that the appropriate proportion of the imposed load that

should be considered as permanent is:1 ≤ 0.1.
 Definition of ‘low’ and ‘high’ frequency
floors for walking activities
1
2z

Response Ratio, R(t)

Response ratio, R(t)

Time, t Time, t

‘Low frequency floor’ ‘High frequency floor’

3.0 Hz 10.0 Hz >10.0 Hz

Resonance from first four harmonic

Impulsive excitation
components of activity frequency
 Weighted peak acceleration response from
walking activities on low frequency floors

 n P0
a j ,n ,m   i ,m  j ,m DR1W
Mm
where i,m is the mode shape amplitude at the point on the floor
where the excitation force is applied, j,m is the mode shape
amplitude at the point where the response is to be calculated, n is
the Fourier coefficient of the nth harmonic, P0 is the static force
exerted by an ‘average person’ (normally taken as 76kg × 9.81m/s²
= 746 N), Mm is the modal mass of mode m, D is the dynamic
magnification factor, R1 is the resonant build-up factor and W is the
appropriate weighting factor given by ISO 2631.
Fourier coefficients to be used for design

Harmonic, n Fourier coefficient, n

1 0.60 fp – 0.41

2 0.20

3 0.10

4 0.05

Note: 1.5 Hz ≤ fp ≤ 2.5 Hz

 Dynamic magnification factor

n 
2 2
D
1  n    2nz 
2 2 2 2

where n is the number of the nth harmonic (n = 1, 2, 3 or 4),  is the

frequency ratio (taken as fp/ fm) and z is the damping ratio.
 Resonance build-up factor

2f mzTw
R1  1  e
where Tw is the walking duration (Tw = D / V) in seconds, D is the
characteristic dimension in metres taken as the longer of the floor’s
plan dimensions or, where known, the corridor length and V is the
walking velocity in m/s given by

2
V  1.67 f p  4.83 f p  4.5
where fp is the pace frequency.
 Paris Office

3'
F loor be a m S 2 1 2 2' 3 4 5 6 7
9450 4700 7450 12150 4050 12150 12150
1350
F 2900
F loor be a m S 1

E
6750

D AR E A 3
AR E A 1
7000

13000 Area 1 (bare)

First mode frequency = 4.13 Hz
B
6750
Area 1 (finished)
A' First mode frequency = 4.19 Hz
6750 AR E A 2
Area 3 (finished)
A First mode frequency = 4.44 Hz
F loor be a m S 1
Area 3 - Predicted frequencies compared
with measured frequencies

10

8
Frequency (Hz)

Predicted
6 Frequency
4 Measured
Frequency
2

0
0 1 2 3 4 5 6 7 8
Mode number
Area 3 – Predicted response factor versus
pace frequency

0.120
Frequency-weighted
acceleration (m/s²)

Harmonic 1
0.080 Harmonic 2
Harmonic 3
Harmonic 4
0.040
Total

0.000
1.5 1.7 1.9 2.1 2.3 2.5
Pace frequency (Hz)
 Weighted peak acceleration response from
walking activities on high frequency floors

I
a j ,m  2f m 1  z  i ,m  j ,m
2
W
Mm
where fm is the frequency of mode m, z is the damping ratio, i,m is
the mode shape amplitude at the point on the floor where the
impulsive force I is applied, j,m is the mode shape amplitude at the
point where the response is to be calculated, I is the Impulsive
force, Mm is the modal mass of mode m and W is the appropriate
weighting factor given by ISO 2631.
 Effective impulse forces from walking on
high frequency floors

1.43
fp P0
I  54 1.3
fm 700

where fp is the pace frequency, fm is the frequency of the mode

under consideration and P0 is the static force exerted by an
‘average person’ (normally taken as 76 kg × 9.81 m/s² = 746 N).
 SCI Headquarters

A B C D E
600 0 600 0 6000 6000

3
7450

2
7450

1
SCI Headquarters – Predicted versus
Measured Floor Frequencies
14
13
Frequency (Hz)

12
11 Predicted
10 M easured
9
8
7
0 1 2 3 4 5 6 7 8 9

M ode Number
SCI Headquarters – Predicted response factor
versus pace frequency

0.040
Frequency-weighted
acceleration (m/s²)

0.030
Reponse due to
0.020 impulsive
excitation

0.010

0.000
1.5 1.7 1.9 2.1 2.3 2.5
Pace frequency (Hz)
Summary of measured floor responses for
8 floors cf. predictions

14
Measured multiplying factor
 = 0.86 Floor 1
12
COV = 23% Floor 2
10
Floor 3
8 Floor 4
6 Floor 5
Floor 6
4
Floor 7
2 Floor 8
0
0 2 4 6 8 10 12 14
Predicted multiplying factor
 Conclusions
A method for calculating the vibration
performance of steel-framed floors has been
presented.
Frequency predictions from FE analysis have
shown reasonable correlation with measured
values
Response predictions, based on the modal
properties from FE analysis, have been
shown to correlate quite well with
measurements from walking tests