T R A N S P O R T A N D ROAD RESEARCH L A B O R A T O R Y

Department of Transport

RESEARCH REPORT 207

THE EFFECTS OF TRAFFIC INDUCED VIBRATIONS ON

HERITAGE BUILDINGS--FURTHER CASE STUDIES

by G R W A I ' r S

The views expressed in this report are not necessarily those of the Department
of Transport

Vehicles and Environment Division

Vehicles Group
Transport and Road Research Laboratory
Crowthorne, Berkshire, RG11 6AU
1989

ISSN 0266-5247
CONTENTS
Page Page

Abstract 1 7.4.1 Building A - - t o w n house 13

1. Introduction 1 7.4.2 Building B--church 14

2. Generation and propagation of traffic 7.4.3 Building C--timber-framed

vibration 1 cottage 14

3. Selection of buildings 2 7.4.4 Building D--farmhouse 14

4. Measurement method 2 7.5 Traffic surveys 14

4.1 Vibration measurements 2 7.6 Soil surveys 15

4.1.1 Instrumentation 2 8. Discussion 15

4.1.2 Structural vibrations 4 8.1 Exposure to vibration 15

4.1.3 Window vibration 4 8.1.1 Radial component of

vibration 15
4.1.4 Crack movements 4
8.1.2 Vertical component of
4.2 Noise measurements 4 vibration 16

4.3 Measurement of road surface 8.2 Window vibration 17

irregularity 4
8.3 Causes of damage 17
5. Surveys 6
9. Conclusions 18
5.1 Building surveys 6
10. Acknowledgements 18
5.2 Traffic surveys 6
11. References 18
5.3 Soil surveys 6
12. Apendix--Reports of the structural
. Analysis of vibration data 6 surveys by the Historic Buildings and
Monuments Commission for England 19
6.1 Time domain 6
12.1 1 and 3 Widcombe Parade, Bath 20
6.2 Frequency domain 6
12.2 The parish church of St James,
7. Results 7 Louth 29

7.1 Vibration levels 7 12.2.1 Comparative study of

windows 29
7.1.1 Vibrations at foundations 7
12.3 The Old Bell Cottage, Norton 35
7.1.2 Amplification of vibrations 9
12.4 Rising Sun farmhouse near Honiton,
7.1.3 Source of vibrations 9 Devon 39

7.1.4 Noise levels 11

7.1.5 Window vibrations 11

7.2 Dynamic crack movements 13

© CROWN COPYRIGHT 1989
7.3 Road surface movements 13 Extracts from the text may be reproduced,
except for commercial purposes, provided the
7.4 Building surveys 13 source is acknowledged.
Ownership of the Transport Research
Laboratory was transferred from the
Department of Transport to a subsidiary of
the Transport Research Foundation on I st
April 1996.

This report has been reproduced by

permission of the Controller of HMSO.
Extracts from the text may be reproduced,
except for commercial purposes, provided
the source is acknowledged.
 THE EFFECTS OF TRAFFIC INDUCED VIBRATIONS ON
HERITAGE BUILDINGS--FURTHER CASE STUDIES

ABSTRACT As in the first phase buildings were sought close to

roads carrying substantial lorry traffic and where
The effects of traffic induced vibrations on heritage there were obvious signs of building distress. In
buildings have been studied in four widely different addition there was a requirement for a site where
buildings. Three stone-built and one timber-framed lorry traffic would increase over the study period.
building, including a Georgian town house and a The objectives were firstly to quantify the vibration
large 15th century church, were examined. As part of exposure of various building elements and determine
the study a large farmhouse was surveyed before and the dominant modes of vibration, secondly to
after a significant increase in lorry traffic to determine determine the condition of the buildings and lastly to
if the increased exposure to traffic vibration had establish as far as was possible the main causes of
triggered damage. This study extends and any observed damage.
complements a previous examination of four
relatively small brick built properties reported in The Historic Buildings and M o n u m e n t Commission
RR156 (Watts 1988). All buildings were within a few for England (English Heritage) who have
metres of roads carrying HGVs and they all showed responsibilities for historic buildings and monuments
some signs of distress. Vibration and noise of national importance collaborated in this research
measurements were made at the sites in order to by carrying out structural surveys of the buildings,
characterize the exposure to vibration. All building monitoring crack movements and reporting on their
surveys were carried out by the Historic Buildings observations and conclusions concerning the effects
and Monuments Commission for England whose of traffic vibration.
main objectives were to report on the state of
dilapidation of each building and to determine
whether any damage was the result of traffic
vibration. It was concluded that although measurable
levels of ground-borne vibrations were present in all
buildings, reaching perceptible levels in some, the 2 GENERATION A N D
observed damage was likely to have been caused by
other site factors rather than the exposure to traffic PROPAGATION OF TRAFFIC
vibration. VIBRATION
Passing vehicles can induce building vibration in t w o
major ways. Low frequency noise produced by
1 INTRODUCTION vehicle engines has dominant frequencies in the
50-100 Hz range and enters buildings readily through
Residents living alongside heavily trafficked roads are doors and windows. It can produce vibrations in
often concerned about the damaging effects of building elements particularly if they are light and
traffic vibration on their homes (Watts, 1984). There flexible. High levels of vibration can be measured on
is also a widespread belief that old and structurally w i n d o w panes fronting heavily trafficked roads and
weak buildings are particularly at risk from these this can produce annoying rattles (Wa~s, 1984). At
vibrations (Civic Trust, 1970; Crockett, 1973). A the most exposed locations, acoustically induced
previous report (Watts, 1988) described the first floor vibrations can become perceptible but vibration
phase of an investigation into the possible link levels in the hard structure of the building are
between traffic vibration and building damage. Four generally at a much lower level (Watts, 1987b).
grade II listed buildings were studied and all were
brick built properties used wholly or partly as private Ground-borne vibration has dominant frequencies in
dwellings. In two cases control properties of similar a lower frequency band, typically 8 - 2 0 Hz. These
construction and age were also studied in order to vibrations are produced by the varying forces
aid the assessment of the causes of damage. It was between tyre and road and become perceptible in
concluded that the observed damage could more buildings near the road if the axle loads are high and
plausibly be attributed to factors other than traffic the surface is uneven. Both compression and shear
vibration. In this second phase a much wider range waves are produced and their amplitudes and
of building types in terms of age, size and attenuation with distance depend on a number of
construction was selected for study and one of the factors including the soil composition and the nature
buildings was monitored before and after a of geological layers beneath the surface. Since this
substantial increase in lorry traffic. In these ways it vibration enters bulidings through the foundations,
was expected that further insights would be gained the hard structure of the building is normally affected
into the role of traffic vibration in causing damage to to a greater degree than is the case for airborne
buildings of widely different types. vibrations.
There are two further mechanisms for the generation probably as a private dwelling house but later
of traffic induced vibration that need to be converted to shop premises on the groundfloor. It is
considered where buildings are situated very close to situated 3 m from the A36 which is a two lane one
the carriageway. Firstly, the air pressure pulses way system at this point. Building B is the parish
caused by the bow and stern waves of large vehicles church of St James in Louth in Lincolnshire. It is a
passing at speed and secondly the quasi-static very large building dating from 1430; the spire was
depression of the road surface under the weight of added in the 16th century. The east wall containing
heavy vehicle axles may induce small, short duration large stained glass windows adjoins the A16 and the
vibrations or movements at the front wall. The buttresses are less than 1 m from the carriageway
frequency of any vibration would be expected to be (see Plate 2). The road is so narrow at this point that
related to the length of the vehicle and its speed and traffic lights control vehicle flow in one direction at a
probably would not exceed 3 or 4 Hz in most cases. time. Building C forms the south end of a T-shaped
At one site where site conditions were considered to group of timber framed buildings at Norton, near
be conducive to observing these movements, further Evesham, and is basically a rectangular two storey
measurements and analyses were carried out to dwelling. It was probably built in the 15th century
indicate the scale of the effects. with the porch and bell cote added in the 19th
century. There is a mid 20th century two storey
extension at the rear. The east gable wall is timber
framed with lath and plaster infill panels and is
situated only 2 m from the heavily trafficked A435
3 SELECTION OF BUILDINGS (see Plate 3). Building D stands at a crossroads on
an unclassified road about one kilometre south of the
One aim of this second phase of study was to A30 leading to Honiton. It was built as the 'Rising
determine the effect of traffic induced vibrations on a Sun' inn at the beginning of the century and has
greater range of building types and this was achieved since been converted to a farmhouse. It is a two
by including a timber-framed dwelling and three storey detached rectangular building with a front
stone built structures of widely different ages and entrance porch only 3.4 m from the edge of the
proportions. Although it was not possible to find carriageway. The external walls are constructed
similar buildings not exposed to traffic vibration at
mainly from flint and stone. Plate 4 shows a pipe
each site to aid the interpretation of observed
delivery lorry passing close to the building, at a gross
damage, it was possible to study the effects of
weight estimated to be over thirty tonnes.
vibration on a t w o storey building before and after a
large increase in lorry traffic. In this way any short-
term damage triggered by passing traffic would be
readily detected. In addition, because one of the
buildings was very large (a church over 60 m long) it
was feasible to study damage to parts of the building 4 MEASUREMENT METHOD
which were exposed to different levels of vibration. To determine the exposure of the buildings to
This was particularly relevant to the study of vibration, peak amplitudes and rms values of particle
vibration damage to the stained glass windows which velocity and acceleration were measured at various
at the east end were exposed to significantly high parts of the building. Where possible dynamic crack
levels of traffic vibration. movements were recorded to supplement this
information. The frequency content of the vibrations
A t three sites the buildings selected were all within a
was also computed to indicate the dominant source
f e w metres of roads carrying high lorry flows and
of vibration. Noise measurements were made at the
traffic vibrations were perceptible at the kerbside or
facades of the buildings to quantify the level of low
inside the buildings so that exposure was known to
frequency noise, and measurements were made of
be relatively high before measurements commenced.
the profiles of significant road surface irregularities
A t the fourth site the building was situated close to a
which were considered to be chiefly responsible for
quiet country lane carrying only a few two axled
heavy vehicles per hour and occasionally an ground-borne vibrations.
articulated goods vehicle. Although vibrations at this
site were not perceptible it was expected that 4.1 V I B R A T I O N MEASUREMENTS
following the opening of a gas pipeline store within
4.1.1 I n s t r u m e n t a t i o n
100 m of the house there would be a large growth in
the numbers of very heavy vehicles passing the site Peak particle velocities (PPVs) of major building
resulting in increased exposure. elements were measured in three orthogonal
directions. Long travel geophones were employed
Full descriptions of the buildings and their state of with a sensitivity of 28mV per m m / s and with a level
dilapidation are given in the building survey reports in frequency response down to 5 Hz. The geophones
the Appendix and Figure 1 shows simplified plans for were bolted to the sides of an aluminium cube and in
each building. Building A is a four storey Georgian most cases attached to the measurement surfaces
terraced town house situated in Widcombe Parade, with Plaster of Paris. The signals from the geophones
Bath (see Plate 1). It was built in stone c1760 were suitably conditioned by operational amplifiers

2
 Neg.No. R688 87/4 Neg.No. R739 87/1

Plate 1 Georgian t o w n houses in Bath. Plate 2 East w a l l of c h u r c h at Louth (Site B)

M e a s u r e m e n t s taken in second house
f r o m far end of terrace (Site A)

Neg.No. R782 87 8 Neg.No. R600/88/6

Plate 3 T i m b e r - f r a m e d cottage at Norton, near Plate 4 F a r m h o u s e w i t h p i p e d e l i v e r y l o r r y

Evesham (Site C) passing close to f r o n t porch (Site D)
which provided the input to an intelligent interface were inaccessible, being high up 'in the chancel
unit (CED 1401) driven by a microcomputer which arches. The measurements were made with a strain
was programmed to calculate PPVs on up to 16 gauge which could resolve movements of the order
channels and could provide printed time histories if of one micron. The measurement points are indicated
required. It was also capable of sampling by the l e t t e r ' S ' in Figure 1. A bridge and operational
continuously on one channel and computing the amplifier were used to condition the signals and
number of vibration events which exceeded given simultaneous measurements were made with the
levels. Because of space constraints at site C it was vertical vibration of the front foundations. A t the
not possible to operate the equipment indoors and so town house (site A) measurements were made across
an FM instrumentation tape recorder, set up in an a vertical crack in a party wall on the first floor of
estate car, was used to acquire these data. This the building near the front facade. At the timber-
recorder was also used at other sites to record framed cottage (site C) the measurements were
vibrations continuously during 15 min periods so that taken across a crack in an outside wall on the
rms levels could be later calculated. A t site D, where groundfloor again near the front facade.
vehicle f l o w was low, the rms value was very small Measurements were taken above the porch door
and so only peak values were recorded as individual fronting the road at the farmhouse (building D).
heavy vehicles passed by.

4.1.2 S t r u c t u r a l vibrations 4.2 N O I S E M E A S U R E M E N T S

A t each building, measurements were taken at A t buildings A and C, external low frequency noise
foundation level on the front wall close to the major measurements were made 1 m from the front facade
source of ground-borne vibration (position 1 in Figure at first floor level (see letter 'M' in figure 1). At the
1) which was either a poorly backfilled trench, church (site B), measurements were made at a
sunken drain cover or patch of repaired road surface. height of approximately 3 m above ground level in
Simultaneous measurements were taken at various front of the east and west wall windows. Traffic
levels on the front and rear wall so the overall levels were very low past the farmhouse (site D) and
response of the building could be determined and the although measurements were attempted they were
amplifications with respect to the levels at the front found to be contaminated with the noise from the
foundations could be computed (see points 1 - 5 in construction plant operating in the pipe store less
Figure 1 for the principal measuring positions). than 100 m away and for this reason recordings were
Vibrations at the front foundations were monitored not made. All measurements were made with a ½
for 15 min period each hour throughout the day so inch condenser microphone connected to a
that the daily vibration dose could be characterised. measuring amplifier which provided the polarization
voltage. A 22 Hz high pass filter was incorporated so
that low frequency surges produced by air turbulance
4.1.3 W i n d o w vibrations would be reduced.
Short term measurements of w i n d o w vibration were
made at sites A, B and C with a small piezoelectric
accelerometer (sensitivity 3.75 mV per m/s2). The
accelerometer was attached with beeswax to the 4.3 M E A S U R E M E N T OF R O A D
middle of a w i n d o w pane on the exposed facade of SURFACE IRREGULARITY
the building. A measuring amplifier was used to
condition the signal and recordings were made on A laser plane was used to measure variations in
the instrumentation recorder. Additional recordings height near wheel tracks at each site where it was
were made at the church (site B) since a number of evident that an irregularity was contributing
the stained glass windows rattled and buzzed as significantly to the measured vibration levels. At the
vehicles passed by and there was concern that these farmhouse (site D) measurements were made over an
windows might have suffered damage. identical profile in the before and after period to
Measurements were made on t w o stained glass check whether the large increase in the number of
windows near the exposed south east corner and on very heavy vehicles had caused any further
t w o very similar windows near the south west corner deterioration of the road suface which might in turn
which were not so exposed to traffic vibration in lead to increased vibration amplitudes at the building.
order that any excess damage due to high levels of The laser plane consists of a portable battery
traffic vibration might be detected (Figure 1 shows operated laser source and measuring rod. The source
the measurement positions). was placed at the side of the road and emitted a
rotating beam of laser light in a horizontal plane. The
measuring rod was usually placed at 100 mm centres
4.1.4 Crack m o v e m e n t s along the profile which was defined by a strip of
Dynamic movements of cracks can occur as vibration white tape. Height readings were taken at each point
propagates through building elements. These by holding the rod vertically and activating a moving
movements were recorded at sites A, C and D but detector which came to rest in the plane of the laser
not in the church (site B) since the principal cracks light.

4
 :i...." ..~....':.'.. . ~ ,. ~ .:.......-.......-,'.....,....
E
~ici:(i.:..-.;......-i...i :.. :. ." • • .
.. . o
~ O~
~ .~ .~_
Eoo

°1 _~EE 0

I
I
~ E E . o_

Z ~

g I ~ ~ c~ ~r i -
1 0
"
e-
0

E
EL

o E
I
E
0
¢,J

I o "0
C

C
0

£
E
QI

0 ~- Do
E
.'- ~

0
C
< o~ o~ 0¢-

~CN 0
!

r-
! I

L
0

0
0
..J

0
O o
m I--
E
it

C~
. • ...'. "'o;
• -...-

"':'g ~O
I .".'.::iN <
I
1,
",° J
5 SURVEYS 6 ANALYSIS OF VIBRATION
In addition to the measurement of vibration and
DATA
noise, surveys of the buildings, traffic and soil were Time and frequency domain analyses were used to
completed. analyse the data. Most of the time domain analyses
were carried out on site.

5.1 B U I L D I N G SURVEYS 6.1 TIME D O M A I N

Building surveys were carried out by structural The CED 1401 intelligent interface unit, which was
engineers from English Heritage at the same time or connected to the geophones and strain gauges as
shortly after measurements were made. Their detailed above, was programmed to carry out time
objectives were to describe the state of dilapidation domain analyses as follows:
of the buildings, for example the degree of cracking,
the extent of anydistortions in building elements and (i) The vibration dose at the foundations recorded
any settlement problems and then to infer the most over each 15 rain was divided into 8s intervals
likely causes of this damage based on available site and the number of intervals during which the
evidence and also on a knowledge of problems in PPV in the verticle direction exceeded levels of
buildings of similar type and age. The positions of 0.14 mm/s and 0.3 mm/s was logged. The
significant cracks were marked on drawings and at figure of 0.14 mrn/s corresponds to the base
the farmhouse (site D) small studs were placed curve specified in BS 6472 (BSI, 1984) below
across these cracks and width readings taken with a which comments or complaints about vibration
Demec gauge. Several measurements were taken in buildings are rare. The value of 0.3 mm/s
over a period of months to determine if the increased corresponds to the threshold of perception for
heavy traffic had caused deterioration. The complete continuous sinusoidal vibrations established by
reports of the surveys are reproduced in the Reiher and Meister (Steffens, 1974). The
Appendix. highest vibration amplitudes in the 15 rain
periods were also recorded. Because of the
small number of lorries at site D the equipment
was triggered manually for each passing HGV
5.2 TRAFFIC SURVEYS and the peak values were established in hourly
The volume and composition of the traffic at a site is intervals and not during 15 rain periods.
one of the important factors determining the rms and (ii) The amplification of ground-borne vibrations in
peak levels of vibration recorded in a building. At various parts of the structure was determined
sites A, B and C estimates of the daily variation were by calculating the ratios of the PPVs of major
obtained by recording vehicle flows during 15 rain building elements to the corresponding PPVs
periods each hour between 9:00 or 10:00 and 19:00 of vertical vibration at the front foundation.
hours. Because of low traffic flow at site D, Peak amplitudes were captured for several of
measurements were taken over complete hours the largest vibration events at each
during the before and after periods. Vehicles were measurement position and the average
categorised into 'light' (cars and small vans with a amplifications calculated.
gross weight less than 1.5 tonnes), 2, 3, 4 and 5 or
more axle goods vehicles greater than 1.5 tonnes, (iii) Up to 12 channels of data from four separated
and buses and coaches. Additional information on geophone arrays at the front and rear of the
traffic levels in past years was provided by the local building were plotted to indicate the significant
authority. modes of vibration.
(iv) Recordings of the free vibration of suspended
floors produced by a small vertical impulse
were plotted to determine the natural
5.3 SOIL S U R V E Y S
frequency of oscillation.
The aim of these surveys was to provide a simple
description of the soil type and its moisture content (v) The maximum amplitudes of window vibration
in order to confirm the description obtained from and dynamic crack movements were
large scale drift deposit maps. Where possible the computed.
findings were supplemented by detailed information
on geological strata in the vicinity, obtained from 6.2 FREQUENCY D O M A I N
borehole data held by the British Geological Survey.
Soil samples were taken at various depths in the A two channel digital signal analyser connected to a
range 0.5 to 0.7 m close to the building foundations plotter was used to compute and display the
and sufficient samples were collected to enable the frequency content of the recorded signals. In most
particle size distribution and moisture content of the cases the analysis was restricted to the range
soil to be established. Standard test procedures were DC-100 Hz and the corresponding resolution was
followed (British Standards Institution, 1975). 0.39 Hz. The spectral densities of vibrations at the
 TABLE 1
Vertical vibration levels at foundations by site and time of day

~ ite Number of events in 15 min

period when peak velocity
Number of events in 15 rain
period when peak velocity
Hour ~ >0.14 m m / s >0.3 m m / s Peak level during period
b e g i n n i n ~g ~ At B C A B C A B C DI* D2*

10:00 27 0 18 8 0 6 0.92 0.13 0.42 0.02 0.12

11:00 31 1 25 12 0 8 0.46 0.26 0.44 0.05 0.14
12:00 22 1 17 6 1 4 0.92 0.30 0.45 0.07 0.12
13:00 22 0 26 8 0 9 0.61 0.10 0.46 0.09 0.16
14:00 26 0 22 10 0 2 0.54 0.08 0.37 0.11 0.13
15:00 29 2 16 5 1 5 0.44 0.33 0.37 0.09 0.16
16:00 21 0 19 6 0 2 0.67 0.12 0.41 0.07 0.13
17:00 20 0 9 2 0 3 0.36 0.11 0.39 0.06 0.16
18:00 16 ,2 17 7 0 1 0.49 0.19 0.37 0.09 --
tSite A: town house in Bath.
Site B: church in Louth.
Site C: timber-framed cottage near Evesham.
Site D: farmhouse near Honiton.
* Because of small numbers of HGVs, peak values in hourly periods were recorded.
D~ are results for the period before the increase in lorry traffic.
02 are results for the first period when lorry traffic was relatively high.

foundations, front facade, and middle of suspended lists these values for 15 rain sampling periods
floors and window panes during 15 min periods were throughout the day at sites A, B and C. A t the
determined using signal averaging *. The sound farmhouse (site D), peak values in a full hour of
pressure at facades was also analysed in a similar monitoring are given both for the period before the
way. For the vibration records this enabled the increase in traffic and during the period when lorries
dominant frequencies to be identified and rms values were delivering the pipes to the store. The highest
in the low (4-20 Hz) and high (20-100 Hz) frequency vibration amplitude of 0.92 m m / s was recorded at
bands to be determined by integration of this spectral the town house (site A) and the frequencies with
density function. In the case of the noise recording, which the threshold was exceeded at this site also
the decibel scale was used and integration yielded tended to be greater than at other buildings. An
the dB level in the 20-100 Hz band. examination of the table shows that there is no
obvious systematic pattern in exposure throughout
the day. A t building D the significant increase in
peak levels in the after period can be clearly seen in
each hour. Figure 2 shows this change in vibration
7 RESULTS dose in greater detail. Only one event produced a
A large quantity of data was collected and analysed PPV greater than 0.1 m m / s in the before period yet
and the most important results are presented below. in the after period 32 exceedences occurred in a
similar period. These increases were almost entirely
due to the pipe delivery lorries.
7.1 V I B R A T I O N LEVELS
7.1.1 Vibrations at foundations The frequency contents of the vertical vibrations at
the front foundations are shown in Figure 3 where
The vibration dose at each building was characterised the spectral density is given for sites A, B and C
by the frequency with which perception levels were during a 15 min sampling period. It is clear that at all
exceeded at the foundations in the vertical direction sites the dominant frequencies are below 20 Hz and
and by the maximum recorded amplitudes. Table 1 the peaks of the distribution are in a narrow range
from 9 - 1 2 Hz. By integrating the spectral density
function and taking the square root, the rms value in
m m / s was calculated for low and high frequency
* The spectral density referred to is the autospectrum and is bands (see Table 2) and this again indicates the
the magnitude squared of the Fourier transform of the time dominance of low frequency vibrations. The rms
signal. It indicates the distribution of power in the vibrations
across the selected frequency band. levels are very low at the church (site B) since it was
A linear average of several hundred spectra was computed only occasionally that heavy vehicles crossed a small
for each analysis period. sunken drain cover producing perceptible vibrations
 30
28
26

22
20 ~
K)t
.Q
>0( ~ Before period
16 =-
"5
14 :::::: "_ "_ ~ HGVs delivering pipes 1
!ii!ii ~ and sand to store ( After
12 :::!::: ~ ~ ~ Other HGVs I periOd
u_

8 . .

6 . .

4 . .

o
o-o.o49 0.05-0.099 o.1 - o . 1 4 9 3o.15
Peak velocity of event (ram/s)
measured near foundations of porch

Fig.2 Frequency of vibration e v e n t s by peak velocity produced by HGVs at

Site D in the before and after periods from 9:00--18:00 hours

(ram/s) 2 per HZ

0.2x10 - 3

0.1x10 - 3

0 50 100

0.01x10 - 3 ,

Site B -- Church

0.005x10 - 3 w

0 i
o
_i . . . . . . . 1 .....

5O
L ...... l
k
_,.J~dld'lll~ .J .... I~
lOO

0.06x10 - 3 ,

Site C -- Timber-framed cottage

0.03x10 - 3

_ ~J~.~ I I -J--- I I
0 ,h
0 50 100

Frequency (Hz)

Fig.3 Spectral densities f o r t h e vertical c o m p o n e n t s of particle velocity at the f r o n t f o u n d a t i o n s

8
 TABLE 2
RMS levels for vertical vibration at foundation level

RMS velocity

4-20 Hz 20-100 Hz Frequency at peak

Site Description (mm/s) (ram/s) (Hz)

A Town house in Bath 0.027 0.022 9.4

B Church in Louth 0.005 0.002* 11.3
C Timber-framed cottage near Evesham 0.015 0.010 10.9

* Value estimated due to extraneous measurement no=se.

at the foundations. Because of this, extraneous noise movement of the east and west walls in the radial
was a problem and resulted in the spurious peaks direction even at a height of 30 m in the tower at the
between 60 and 80 Hz. It is estimated that the rms west end of the church. In contrast there was
value for the higher frequency band was inflated by relatively large amplification in the radial direction at
about 40 per cent and the corrected value is given in first floor level at sites C and D. There was little
the table. attenuation of this motion at the rear of the
farmhouse (site D) at first floor level, and both walls
appeared to rock in phase on their foundations.
7.1.2 A m p l i f i c a t i o n o f v i b r a t i o n s
There was however a significant reduction in the
The amplifications of the vibrations that occurred in radial motion at this level at the timber-framed
the buildings are shown in Table 3 where the PPVs cottage (site C), since the amplification reduced from
at various points in the building are related to that 3.4 to 1.2
measured in the vertical direction at the front
foundations. The numbers in brackets refer to the
measurement points shown in Figure 1. Radial and 7.1.3 S o u r c e o f v i b r a t i o n s
transverse measurement directions are also defined in Building C was only 2 m from a de-restricted stretch
this Figure. In all cases at the front foundations the of road where HGVs frequently passed at speed and
vertical component was largest. Vibrations of the it was thought that there was a possibility that
front facade tended to be greater at higher levels and detectable low frequency vibrations due to air
this was especially true of vibrations in the radial pressure pulses or quasi-static depression of the road
direction where amplifications of over four occurred. surface might be evident. As explained in section 2,
On the rear facade vibrations were lower but the the frequency of these vibrations would probably not
increase with height was still apparent. Vertical exceed 3 or 4 Hz in most cases. To examine this
vibrations of suspended floors were in all cases much possibility, a spectral density plot of the radial
higher than at the front foundations, the component of velocity at the front wall at first floor
amplifications in this case ranging from 2.5 to 5.1. level was computed from the average of eight of the
The increase with height is well illustrated at the most significant vibration events caused by HGVs
town house (site A) where amplification increases passing close to the building. This is shown in Figure
from 0.9 on the ground floor which was solid, to 5.1 4, and even allowing for the reduction in sensitivity
on the suspended wooden floor at third floor level. of the geophone below 5 Hz (eg sensitivity at 2 Hz is
Table 4 lists the natural frequencies and dimensions approximately -~ that at 10 Hz) there is no evidence
of these suspended floors together with the dominant of any substantial vibrations at these lower
frequency of vertical vibration at the foundations. frequencies which might result from these
mechanisms.
A print out of particle velocity at several positions on
the front facade at site A showed the principal mode To demonstrate that the source of the observed
of vibration was a rocking motion on the foundations vibrations at this site was due to ground-borne
at about 11 Hz since the vibrations at various heights vibrations, simultaneous measurements in the radial
were approximately in phase and the radial motion at direction were made at the kerbside and on the front
top floor level was nearly four times the wall at first floor level. Time histories of the
corresponding level at the foundations. The church vibrations for the period when a large three-axled
(site B) was over 60 m long which meant that most lorry passed close to the building at speed are given
components at the west end were much reduced due in Figure 5. By close inspection it can be seen that
to the attenuation with distance. The time delay for the first six major oscillations in the vibration traces
vibrations to reach the far wall was 0.34s (wave for the kerb and wall geophone are similar. The
speed 180 m/s) and opposite walls were vibrating out dominant vibrations in both records occur at a
of phase as a consequence. There was relatively little frequency of 11 Hz and cross-correlation analysis
 CO ~ ~ I~ 1~
~ o ~
d om d d

0 0~1 ~ ~ ~ 1.0

dE~ < ,,-

mo~: ~ ~o

d d ~ d d
0

>.
t- 0
C
O ~D

0
_I
.2
>-
d od d d c- r- "1- ~ 0 0
°0_ ~.)
r- 0
~: "I0 0 O"
ffl
d oo d d o m
,9
° ¢- "o
~O 1.00~ 0 Cxl
> d od d d
13_
o
"13- I o
N~
m z t-
o
(1)
"0
0
0
~'0 0
e-
0 'o

0 dddd - O"
~n I ,~ ~0 ~0 "e-
o
o
o ~--r"
Q.
0 0 0 0 ~ "o
o ffl

0
IJJ Z
-,I Q. Z
..J •" ~ 0
rn
Z
o ~ ~ ~
0 ~~ 0 ~
°_
._ i i
(/)
~ g o
.0
_ 0
.o_ '~ EEEE E o
(-..
(1) A "o
> EE
N ddd d~o X X X X o
0
0 .~ ~
I~_ "o o 0 0") O~ ~ ~ ',d" ~ (~0
o C
~d~ d ~E ~ ~: > ,-r o
0
°o g X C~ 0 exl 0~__ ~ ~ c~ "8
~-. r. co o~.~ ~o co co o o
e~
dd~ d~ ~ ~ ~i 0 "o
> " I
o
.E
0 c-

~., t E~ O
o
0
>.
co
,.j 0
0 E
< 0
._J •- - ¢N ('o , r - , -
0 cr
0
_J ,-_m - -

e
d dd d d d
M
0
"~ E
N._o d do d d d
0 0 '~" 0 0
0 0~"~ 0 0
o
e- 0
0
u. I
E
(/)
0 '~
_c L o
gog ~- CD.C
0 E
O') .~- ~ 0 .-- CO o
t-
~ E ~e Z
O

10
 0.2
Based on the average of eight spectra

O.

0.1
E
v
E

f ~......j,'L~ _ i I I I .I I
50 100
Frequency (Hz)
Fig.4 Spectral density for the radial component of particle velocity of
the front wall at Site C at first floor level

0.2
Radial velocity at kerb

0.1 m

E
E

-0.1

--0.2 I I I I I I I I
1 2 3 4

0.8
Radial velocity at first floor level on front facade

0.4

E
E

--04

--08 I ! J ! ! I J i I I
1 2 3 4

Fig.5 Time histories of vibration during the passage of an HGV at SiteC

shows a delay of 80 ms. It would appear that the on flexible structures such as suspended floors
wall is lightly damped and continues to vibrate at a (Watts, 1987b) the rms values for floor vibration
frequency close to the forcing frequency after the given in the table indicate the importance of
major vibrations at the kerb have decayed. It is also ground-borne rather than airborne vibration as levels
apparent that the wall vibration is later re-excited by in the l o w frequency band 4 - 2 0 Hz are greater than
minor peaks of ground-borne vibration. in the high frequency band where acoustically
coupled vibrations normally tend to dominate.
7.1.4 Noise levels
Table 5 lists the linear noise levels recorded at the 7.1.5 Window vibrations
building facades in the 20-100 Hz band and the In the case of w i n d o w vibration the d o m i n a n t
frequencies at the peaks of the spectra. Although frequencies are much higher and peak acceleration
low frequency noise produces the greatest response amplitudes exceed 10 m / s 2 (Table 6). A t site B,

11
 TABLE 5
Noise levels at facades and upper floor vibrations

Noise Vibration

RMS vertical floor

Linear levels at facade (dB) velocit' (mm/s)
Frequency at Frequency at
Position 20-100 Hz peak (Hz) 4-20 Hz 20-100 Hz peak
Site

First floor 89.2 59.2 0.106 0.027 15.2

A
(Town house)
B East side 85.4 40.6
(Church) West side 74.0 35.0

Both measurement ~oints at

a height of 3 m
First floor 86.5 76.8 0.044 0.029 15.6
(Timber-framed
cottage)

At side D, due to low traffic flows, rms vibration levels and noise levels were both very low and affected by
extraneous factors.

TABLE 6
Maximum recorded window vibrations

Peak acceleration Dominant frequency

Dimensions of
Site Position of window window pane (m/s 2) (Hz)

A First floor fronting road 0.34 x 0.27 m 10.4 200

(Town house)
B
(Church)
Stained glass window near
SE corner exposed to traffic
1.95 x 0.55 m ~ 5.7 (traffic)
3.5 (organ playing)
39.9
42.1
Very similar windows near 1.95 x 0.55 m 1.4 (traffic) 43.8
SW corner--not exposed
Ground floor fronting road 0.78 x 0.24 m 5.6 138.9
C
(Timber-framed
cottage)

Recordings not taken at site D because of adverse weather conditions.

TABLE 7
RMS window acceleration at church (site B)

RMS acceleration

Window pane 4-20 Hz 20-100 Hz Linear noise level (dB)

Position of windows dimensions (m) m/s 2 m/s 2 near window

East wall--exposed 2.30 x 0.47 0.012 0.231 85.4

West w a l l - - n o t exposed 2.30 x 0.47 0.011 0.055 74.0

South wall--exposed 1.95x0.55 0.021 0.393

(near east end)
South w a l l - - n o t exposed 1.95 x 0.55 0.004 0.078
(near west end)

12
 measurements were made when the organist played distance to the building was over 30 m. It was
and traffic levels were low in order to provide a thought that this profile might deepen as a result of
context for judging the importance of traffic vibration the large increase in lorry traffic and this might
exposure compared with that resulting from the contribute to higher vibration levels in the after
normal use of the building. The organist was period. Figure 6 shows the profile in the before and
instructed to play the full range of notes in order to after periods and it can be seen that differences are
excite resonances in the stained glass window panes. very small being of the order of a few millimetres at
In this way it was hoped to create the worst case maximum and may in part be due to measurement
conditions likely to be produced by the organ. The error. It was considered that the measured
Table shows that the maximum vibration amplitude irregularities at all the sites produced the largest
produced by the organ was lower than that produced contributions to the vibration levels observed but it is
by traffic although of the same order. Since a likely that other surface irregularities in the vicinity
detailed study of the conditions of similar windows would have made some contribution. This was
exposed to and removed from traffic was made at probably the case at sites A and D where the road
site B a more detailed examination of the vibration surfaces were generally uneven.
dose at two pairs of windows was made. Table 7
lists the rms levels of these four window panes and
also records the linear noise levels at 1 m from the Direction of traffic
exterior of the windows at opposite ends of the 10
church. The window on the south wall at the east o
end 'buzzed' as vehicles passed by and it was clearly
-10
~ Beforeincreasedlorryflow f
driven by low frequency noise. The rms levels are
E -20 ~" " - ~ _ . ., /
substantially greater in the higher frequency band as
can be seen in the table. These levels for the E-
-30
exposed windows are 4 - 5 times greater than the
--40
similar windows away from traffic reflecting the large
differences in sound pressure levels. --50 ~After increas~ed
lorryflow
-60 I I I I I
0 1 2 3 4 5
7.2 D Y N A M I C CRACK M O V E M E N T S Distance along road (m)
At all sites the maximum recorded crack movements Fig.6 Road surface profile contributing to
produced by passing heavy vehicles were very small significant vibrations at Site D
and of the order of one micron, and were in most
cases hardly distinguishable from the background
noise level of the instrumentation. At site D, where
the strain gauge was attached across a crack in the
brickwork directly above the porch, slamming the 7.4 BUILDING S U R V E Y S
front door produced a peak amplitude of 22 microns. The reports on the structural surveys carried out by
engineers from English Heritage are given in the
Appendix. Summaries of the important findings are
7.3 ROAD SURFACE PROFILE given below.
The maximum depths and heights of significant road
surface irregularities at the sites are given in Table 8.
The largest depression at any of the sites was 50 mm 7.4.1 Building A--town house
deep and was due to the presence of a misaligned The building at 3 Widcombe Parade had been altered
service cover and a poorly backfilled trench. However at various stages in its history and this affected the
this did not result in the largest peak levels being ground floor and first floor levels, The only visible
recorded at this site (D) since the propagation sign of distress in the interior was a crack in the

TABL r 8

Characteristics of road surface irregularities

Maximum depth Maximum height Distance from foundation measurement position

Site (mm) (mm) (m)

A (Town house) 5 9 4.9

B (Church) 32 4 6.5
C (Timber-framed cottage) 9 3 2.9
D (Farmhouse) 50 1 33.0

13
party wall next to the stairs between first and second 7.4.4 Building D--farmhouse
floors. On the external facade a crack had developed The building near Honiton was fully inspected on
above the beam over the shop front. This followed four occasions. On the 28th January 1988 the first
the joints in the ashlar blocks and was probably survey was carried out before the increase in lorry
caused by the sagging of the beam spanning the traffic. It was next surveyed on the 20th April which
shop front. The front wall was reasonably plumb, the occurred in the middle of a six week period when the
variation from the vertical up to a height of 5.8 m pipes were being delivered to the store and then
was only 22 ram. again on the 17th June in the middle of a similar
period when the pipes were being hauled back past
the house to be delivered to various parts of the
7.4.2 Building B--church
pipeline. Inspections were also carried out on the
The church was in very good condition structurally. 12th August 1988 and 9th February 1989 after
There were only two lines of fracture that were deliveries had finished. With the exception of the
noted as being of any importance. These were at the porch, the main building was found to be in good
crown of the arches of the chancel both on the north structural condition. The north gable wall had some
and south side and indicate a slight outward lean of minor cracking and flaking to the render. The front
the east gable wall. It was concluded that this and side walls of the porch had an outward lean of
cracking was probably due to the greater weight of approximately 40 mm over a height of 1.5 m and
the east gable wall compared with the smaller weight there were some cracks on the north elevation above
of the side walls. Very many churches of all ages the porch door. The settlement of the porch and
suffer fractures adjacent to the east gable. The associated cracking had possibly started some
condition of the stained glass windows exposed to considerable time ago and had been caused by
high levels of vibration and those not so exposed inadequate foundations or poor subsoil. Little or no
was very similar. further movements had taken place in the porch over
the monitoring period: for example, the recently
repointed corners between the porch wall and the
7.4.3 Building C--timber-framed cottage main wall had not cracked and measured crack
This building has a timber-framed gable wall which movements were relatively small (the largest being
leans towards the road and has a bulge at first floor 0.09 ram) and were probably due to seasonal
level. However there are no indictions that this changes.
movement is continuing and in fact some distortion
is apparent on an early photograph reproduced in a
7.5 TRAFFIC SURVEYS
book published in the early 1940s (Russel, 1941).
Unfortunately the building appears quite small in this The hourly flows are listed in Table 9. For sites A, B
picture since the photograph was taken from the and C these were estimated from 15 rain counts
other end of the village so it is not possible to infer during each hour. The A435 adjacent to the timber-
the degree of tilt accurately. framed cottage (site C) carried the highest lorry flow

TABLE 9
Traffic flow by site
Hourly flow for hour beginning Percentage
Total flow of heavy vehicles
9 10 11 12 13 14 15 16 17 18 10:00-18:00 (%)
Site

A Light* 908 904 976 972 932 968 1008 1368 1344 1016 6472 10.4
(Town house) Heavy** 108 144 184 108 120 128 140 96 60 48 980

B East Light -- 456 568 504 492 588 580 820 564 460 4572 ~} 12.5
(Church) wall Heavy - - 1 0 8 64 112 108 72 88 64 36 44 652
West Light -- 76 132 108 100 100. 164 104 92 92 876 } 6.4
wall Heavy -- 0 8 8 12 8 8 12 4 0 60
Light -- 672 608 616 512 648 752 956 900 780 5664
C 17.6
(Timber-framed Heavy - - 184 148 140 192 184 136 140 88 136 1212
cottage)

D Before Light 48 40 39 45 42 38 42 56 70 11.4

(Farmhouse) Heavy 6 6 10 4 10 7 5 5 1 m

1st After Light 43 49 44 40 41 50 41 52 71 18.5

Heavy 11 9 16 10 11 11 17 8 6
2nd After Light 65 51 60 64 60 73 79 101 105 m
593 \ 16.6
Heavy 19 17 18 15 13 11 14 18 12 118J

* Light: Cars and goods vehicles ~<1.5 tonnes ** Heavy: Goods vehicles >1.5 tonnes, buses and coaches.

14
 both in terms of the total number and also when are soft layers recorded and for such ground
expressed as a percentage of all vehicles. The conditions other work shows that for a given set of
farmhouse (site D) was exposed to the lowest flow conditions relatively large vibrations can be induced
of HGVs, there being only 48 in the measurement in nearby buildings (Watts, 1989).
period before the pipe store was operational.
However this rose to 118 in the second 'after' period
when pipes were being distributed to various parts of
the pipeline. This was more than in the first "after"
period when pipes were being delivered to store and 8 DISCUSSION
was largely due to the opening of a subcontractors
depot nearby at the beginning of the second 'after' 8.1 E X P O S U R E TO V I B R A T I O N
period. Information obtained from the site office All the buildings were situated within a few metres of
suggested that these increased lorry flows were fairly roads carrying HGVs and they were exposed to a
constant over the two six week periods when pipes range of traffic vibration levels. The principal
were being moved. It is likely that the normal lorry vibration frequencies were below 20 Hz and peaked
flows on this road have not changed very much in in the 10 Hz region indicating that the source was
recent years since flow data for October 1986 ground-borne vibration produced by the dynamic
obtained from the Highway Authority shows a similar loading of the road surface by passing heavy
level (61 in a 12 hour period 7:00 to 19:00). vehicles. This was indicated at the timber-framed
Additional flow information was obtained for the cottage (site C) and has been confirmed at other
other sites. At the town house (site A) the total locations near road surface irregularities (Watts,
flows of light and heavy vehicles between 9:00 and 1987a). At sites A and C these levels were relatively
19:00 hours in February 1982 were 7666 and 981 high and vibration amplitudes exceeded the level of
which are similar to those recorded in this study. At perception (0.3 ram/s) many times a day. The
the church (site B) in 1982 the total flow of heavy presence of soft ground at the town house (site A)
vehicles between 8:00 and 18:00 was 685 indicating may partly explain the relatively high vibration levels.
that this flow had not changed substantially in recent At the church (site B) there were few events which
years. At the timber-framed cottage (site C) the flow exceeded this threshold and this was probably due to
of light vehicles recorded in 1981 between 10:00 and the fact that the small sunken drain cover causing
18:00 was 5238 and the total for heavy vehicles was the high dynamic loading was only struck
1331, again similar to levels recorded in this study. occasionally. At the farmhouse (site D) the exposure
was very low before the opening of a pipe store
close to the building but this increased substantially
when the store became operational although it was
7.6 SOIL SURVEYS still below the exposures at other sites.
The particle size distributions and the moisture
contents of the soil samples taken at each site are
given in Table 10. It is evident that there was a large 8.1.1 Radial c o m p o n e n t of vibration
range of soil types: for example, the soil at the Generally vibration levels in the radial direction
farmhouse (site D) contains a high percentage of (horizontally, away from the point of loading on the
very fine particles consistent with a clay while at the road) were higher on upper levels of the buildings
timber-framed cottage (site C) there is a much larger and this has been found in previous studies (Watts,
percentage of larger particles which could be 1987b and 1989) and probably results from the lack
described as sandy gravel. Borehole data showing of restraint offered by the ground. The predominant
the types and depths of various geological strata are mode of vibration in the dwelling houses was a
given in Table 11. Only at the town house (site A) rocking of the front wall about the foundations. At

TABLE 10
Soil characteristics near building foundations

Particle size distribution

Clay Silt Sand Gravel Cobbles

Site <0.002 mm 0.002-0.06 mm 0.06-2.0 mm 2 - 6 0 mm 60-200 mm Moisture content
(%) (%) (%) (%) (%) (%)

A (Town house) 3 26 47 24 0 38
B (Church) 7 23 43 27 0 24
C (Timber-framed cottage) 12 7 35 46 0 15
D (Farmhouse) 43 27 22 8 0 29

15
 TABLE 11
Description of geological strata obtained from nearest borehole

Distance of borehole
Site Description Thickness (m) Depth (m) from site (m)

A (Town house) Made ground 0.23 0.23 60

Silty sandy clay with some gravel--
mainly firm but with some soft layers 6.48 6.71
Clayey sand with gravel 1.75 8.46
Stiff sandy silty clay 4.42 12.88

B (Church) Marl clay 1.37 1.37 470

Sand and stone 1.52 2.89
Marl clay 1.~ 4.57
Sand and soil 0.46 5.03
Marl clay 11.16 16.19

C (Timber-framed Top soil 0.25 0.25 1700

cottage) Firm to stiff silty sandy clay with
occasional gravel 1.40 1.65
Clayey silty sand with gravel 0.95 2.60
Firm to stiff sandy silty clay 0.40 3.00
Stiff calcareous silty clay with very thin
to medium beds of strong limestone 1.30 4.30

D (Farmhouse) Top soil 0.50 0.50 340

Firm to stiff sandy silty clay 1.50 2.00
Dense cobbles and boulders in matrix
of sandy silty clay with gravel 2.00 4.00

the town house (site A) this effect was probably thought to result from the reduction of dead load
exacerbated by the type of construction since the with height (Marshall and Hood, 1987). Amplification
front wall of ashlar blockwork was only 175 mm thick of the vertical component was very noticeable in the
yet over 8 m high. The side and rear walls were middle of floors at the fronts of the buildings. At the
nearly three times as thick and being stiffer town house (site A) the peak amplitude increased
responded to the ground-borne vibration to a much monotonically with height reaching over five times
smaller extent. The church (site B) had thick stone foundation level on the third floor. The amplification
walls with massive buttresses (see Plate 2) and as of the suspended floors relative to the foundation
expected the radial vibrations of the exposed wall level depends also on the mass and stiffness of the
were relatively small. A t the timber-framed cottage floor. Where the mass is relatively high and stiffness
(site C) there was a relatively large amplification of low, the natural frequency will be closer to typical
the radial component probably due to the flexible frequencies of ground-borne vibration and the
nature of the timber frame which was poorly damped response will tend to be high. Thus at site C where
and also because the natural frequency may have the natural frequency was highest the amplification
been close to the principal frequency of ground- was lowest (2.5) and the greatest response occurred
borne vibration. This was not the case on the rear on the floor with the lowest natural frequency where
wall as this was brick built and therefore more the amplification was 5.1 (site C). This trend was
substantial. A t the farmhouse (site D) both front and also noted in the first series of case studies (Watts,
rear walls vibrated in phase at a relatively high 1988).
amplitude. The absence of thick partition walls on
the first floor and continuous joists running from Peak vertical levels on the rear walls of the buildings
front to back probably contributed to this effect. were less than the corresponding levels at the front
since there was some attenuation with distance
especially at site B where the building was over 60 m
8.1.2 V e r t i c a l c o m p o n e n t of vi b rat i on long. In the absence of dissipative losses in the
There was generally an increase of the vertical underlying ground, the minimum attentuation
component of vibration with height but this was not expected can be calculated from the assumed
so large as the amplification of the radial component. circular spreading of surface waves from the road
This was also noted in a previous study and is surface irregularity. This leads to an attenuation rate

16
 inversely proportional to the square root of the similar cracks in parts of the building not so exposed
distance from the source (Barkan 1962). On this basis to traffic vibration at the time so it cannot be
the expected attenuation to the south west corner is concluded on this evidence alone that heavy traffic
0.31 and can be compared with the average value had contributed to the movement. It should also be
measured at the foundations of 0.25 showing that noted that cyclic thermal movements of building
some small dissipative losses had occurred. materials and components are to be expected
(BRS 1970). For example a 6 m length of
unrestrained dense limestone would expand 1 mm if
8.2 W I N D O W V I B R A T I O N S the temperature increased by 30 Celsius. These
Low frequency noise levels at all sites where movements can be progressive particularly if there is
measurements were taken were relatively high an ingress of debris into these cracks. The stained
ranging from 85 to 89 dB at the exposed facades. glass windows chosen for study were of a very
This was expected as these buildings were close to similar age and construction and although the
carriageways carrying high levels of lorry traffic. exposed windows at the east end were subjected to
Consequently window vibration levels were high at average vibration levels 4 - 5 times that measured
these sites and the maximum acceleration of over near the west end there were no differences in their
10 m/s 2 was recorded at site A where noise levels conditions which could be attributed to the action of
were highest. The dominant frequencies in the traffic vibration. This is despite the fact that the
vibration spectra were in the range 40-200 Hz and window near the traffic on the south wall was found
clearly result from acoustic coupling rather than to have been exposed to a peak acceleration
ground-borne vibrations. These frequencies tended to amplitude of nearly 6 m / s 2 and "buzzed' noticeably as
increase with decreasing window pane size. This is vehicles passed. It is often the case that normal use
probably due to a decrease in natural frequency with of a building can produce vibration levels in parts of
window area (Steffens, 1975). the structure which are similar to, or higher than,
those produced by traffic (Watts, 1987b; Watts,
1988). For example, the peak acceleration of a
8.3 CAUSES OF D A M A G E stained glass w i n d o w produced by the church organ
The maximum amplitude of vibration of a structural approached the level caused by passing traffic and
element of any of the buildings was well below levels bell ringing can produce very high levels in a church
known to cause even minor damage directly. Some steeple. Steffens (1974) records a peak velocity due
minor damage from other vibration sources eg to bell ringing of 17 m m / s at 1.4 Hz at a height of
blasting and pile driving has been found to occur just 26 m in a tower 50 m high. This level is far greater
above a PPV of approximately 10 m m / s (Nelson and than that which can reasonably be expected from
Watts, 1987). However damage may possibly be passing traffic.
triggered in a building element weakened through
other causes and there are the further possibilities A t site C the timber-framed gable wall next to the
that low level vibrations over a period of many years road was highly distorted and responded readily to
may cause damage through material fatigue or ground-borne vibration. However an old photograph
induce differential settlement through soil indicates some distortion was present many years
densification or migration.
ago and before the advent of modern heavy traffic.
In addition a timber-framed building within a few
The most significant damage noted at the town miles of this site but situated away from heavy traffic
house (site A) was probably a result of the alteration showed a similar but more severe distortion (see
from a private dwelling house to a shop. The Appendix !2.3).
deflection of the beam over the shop front due to
the weight of the masonary above probably caused
the crack in the stonework. This is a common A t the farmhouse (site D) the large increase in
occurrence under such circumstances and it is exposure to traffic vibration has not been associated
unlikely that the damage was caused by the action of with any significant changes in the building. Changes
traffic vibration. in the crack widths on the porch and north wall have
been very small, and these may well be due to
At the church (site B) the major cracks were in the seasonal changes in temperature and humidity. Some
chancel arches close to the major road. However as minor cracking on the south gable wall that was
pointed out in the Appendix, cracking at these points noticed during the second after period may have
in similar churches not exposed to such vibration is been caused by substantial maintenance work that
common and is thought to result from the was carried out over the period of the surveys.
movements caused by the greater mass of the east
gable wall compared with the less massive side walls. A t all sites where dynamic crack movements were
Measurements were made of these crack widths by measured there was little or no movement due to
the church architects between 1977 and 1981 (Benny passing vehicles and it is therefore likely that these
1987) and show some movements. The largest cracks are not a direct result of exposure to traffic
increase was 2 mm and the largest decrease was vibration. Domestic activity such as slamming doors
1.4 mm. No control measurements were made of has been shown to produce measurable movements

17
in cracks adjacent to the doors. These movements investigations are similar to the first phase and are as
can be greater than those produced by traffic follows:
vibration (Watts, 1987b; Watts, 1988). (i) A t all sites ground-borne traffic induced vibration
was the most significant source of building
Finally, in case studies of this type it is not possible vibration. When the surface is even, airborne
to exclude the possibility that traffic has in some way vibration dominates, but in the cases studied
contributed to the observed damage and this was a surface defects up to 50 mm deep gave rise to
conclusion of the previous study (Watts, 1988). relatively high levels of ground-borne vibration.
Plausible reasons have been advanced to account for A t two sites levels were relatively high, regularly
the state of dilapidation based largely on the exceeding the level of perception at ground level.
judgement of structural engineers who have
specialized in heritage type buildings. Wide (ii) Maximum vibration amplitudes were greater on
experience of common types of damage in other upper floors and walls at the fronts of the
buildings of similar types to the ones examined in buildings than at foundation level.
this study provided the framework for assessing
(iii) Window pane vibration levels were relatively high
whether vibration effects are likely to be important.
and at one site where stained glass windows
However it remains a possibility that traffic vibration
exposed to high levels of airborne traffic
may exacerbate structural problems that are clearly
vibration were compared with similar windows
due to other causes (eg cracking above the shop
not so exposed, no differences in their condition
w i n d o w due to alteration works at site A) or may be
were found which were attributed to the effects
assisting soil settlement (eg the ground under the
of traffic vibration.
east wall of building B causing it to lean towards the
road). In other studies where tighter control of (iv) Damage surveys carried out by structural
extraneous factors was possible there was no engineers from the Historic Buildings and
evidence that traffic vibration has any significant Monuments Commission for England identified a
effects. In one study where a pair of semi-detached range of defects in the buildings ranging from
houses was exposed to high levels of simulated cracks in plaster finishes to more substantial
traffic vibration at levels which would be intolerable structural damage such as cracked stonework,
to most people (Marshall and Hood 1987) the only distortion of walls and foundation settlement. In
observed damage was minor cracking of plasterwork. all cases, however, it was concluded that the
It was concluded that this damage would probably main causes were likely to have been other site
go unnoticed in a normally decorated house. There factors rather than exposure to traffic vibration.
was no movement of the building despite the fact
that it was built on a fairly loose sand and some
densification of the sand was expected. In another
study the condition of houses on alluvial soils
exposed to traffic vibration was compared with that 10 ACKNOWLEDGEMENTS
of very similar properties in the neighbourhood that
were situated well away from heavy traffic to The work described in this report was carried out in
determine if any excess damage was detectable. It the Vehicles and Environment Division of the
was concluded that there was no evidence of Vehicles Group of the TRRL. The assistance of the
significant differences in the condition of the Historic Buildings and Monuments Commission for
properties that could reasonably be attributed to the England is gratefully acknowledged.
effects of traffic vibration. Thus while it is not
possible to exclude the possibility that traffic
vibration has some effect on the buildings in the
present study it is likely that the observed damage
was mainly due to other site factors. 1 1 REFERENCES
BARKAN, D D (1962). Dynamics of bases and
foundations. McGraw-Hill.
9 CONCLUSIONS BENNY, R (1987). Private communication.
The effect of traffic induced vibrations on four
heritage buildings has been examined. This study BRITISH STANDARDS INSTITUTION (1975).
forms an extension to a previous examination of four Methods of tests for soils for civil engineering
relatively small brick built dwellings. The range of purposes. BS 1377. British Standards Institution,
building considered in this further phase was far London.
greater in terms of age, size and type of
construction. BRITISH STANDARDS INSTITUTION (1984).
Evaluation of human exposure to vibration in
The main findings from an analysis of the large buildings (1 Hz to 80 Hz). BS 6472. British Standards
amount of physical measurements and structural Institution, London.

18
 BUILDING RESEARCH STATION (1970). Cracking in
buildings. Building Research Station digest 75. BRE
12 APPENDIX
Garston, Watford.
REPORTS OF THE S T R U C T U R A L
CIVIC TRUST (1970). Heavy lorries. A Civic Trust S U R V E Y S BY THE H I S T O R I C
report. BUILDINGS AND MONUMENTS
C O M M I S S I O N FOR E N G L A N D
CROCKETT, J H A (1973). The effects of traffic in
older buildings. PTRC course entitled: Environmental Although the immediate physical effects of passing
factors in road design. 4-8th June. traffic in terms of vibration, crack movements and
noise can be measured objectively using a variety of
MARSHALL, C and HOOD, R A (1987). Monitoring instruments, assessment of damage to the structure
the effects of simulated road traffic vibrations on a of buildings is a matter of judgement based on
test house--results. In Traffic vibration and building experience. The conclusions drawn in this report rest
damage--TRRL papers presented at Acoustics '87 largely on the expert judgements of structural
(Edited by G R Watts) Department of Transport engineers who have a wide experience of assessing
TRRL Report RR 146, Transport and Road Research damage in heritage buildings. Since the results of
Laboratory, Crowthorne. their surveys cannot readily be summarised without
neglecting some aspects which might be helpful to
NELSON, P M and WATTS, G R (1987). Traffic the reader, their reports are reproduced in full.
vibration and building damage. In Traffic vibration
and building damage--TRRL papers presented at
Acoustics '87 (Edited by G R Watts) Department of
Transport TRRL Report 146, Transport and Road
Research Laboratory, Crowthorne.

RUSSEL, J (1941). Shakespeare's country.

B T Batsford, North Audley Street, London.

STEFFENS, R J (1974). Structural vibration and

damage. Building Research Station Report, Garston,
Wat-ford.

WATTS, G R (1984). Vibration nuisance from road

traffic--results of a 50 site survey. Department of
Transport TRRL Report LR1119: Transport and Road
Research Laboratory, Crowthorne.

WATTS, G R (1987a). Traffic induced groundborne

vibration in dwellings. Department of Transport TRRL
Report RRI02: Transport and Road Research
Laboratory, Crowthorne.

WATTS, G R (1987b). Response of building elements

to simulated traffic vibrations, in Traffic vibrations
and building damage--TRRL papers presented at
Acoustics '87 (Edited by G R Watts). Department of
Transport TRRL Report 146, Transport and Road
Research Laboratory, Crowthorne.

WATTS, G R (1988). Case studies of the effects of

traffic vibration on heritage buildings. Department of
Transport TRRL Report RR 156: Transport and Road
Research Laboratory, Crowthorne.

WATTS, G R (1989). Groundborne vibration

generated by HGVs--effects of vehicle, road and
ground parameters. Paper presented at Acoustics '89
at the University of Oxford. In proceedings of the
Institute of Acoustics, Edinburgh.

19
 ~c~ .~ ~ o
0"" .~,~
0..~I " ~ "~ ..~
~.~ -uo =

o~
0 o
.,~I ~* ~ ~ 0
z~
0 ~

~.~-~
4~
~N

,=,og N

"F. ° "~'' ' ~ ~ ~ ~' ~ ~~ 0~ ~~:" ~ ~ ~

"7. ~' ~ o
.o .
,
o..°o o~_ ~, ~
~.~ ~'~_~ ~ = ~

~_~-6 ~ ~'~ ~ ~ " ~ ~ ~ ~ . ~ .', ~ ~ 0

~'~ ~ "
~= _~R.~'~,~ .~.~ ~ -~ = .~.

-~

~,, ~ ~'~ ~ 0

.n N ~ o ~, ~ Z ~ - '~ ~ ' ~ ~ ' ~ = = ~

o ~ Z

20
 0-~ ~0 ~ 0 "~"~ ~
..= = ~'~ =~
~ ~.= ~
o ~
Q'~ 0 ~ ~,j CO ~a o o o~ - .~'~1

.= "~ •

= ~ E.~ o°== 0 = ~.~

~ ~ 0 .-I= -
o-a "" oZ •~ o .=.,
o. ~ "c~ ~: ~
. v ~ =~ .=.

~2 CO~ 0 ~ ""~

-= =
0
0
0 o
~ ~ .
o ~- ,.5 ..u -~ .~.~ -~ .~
_ ~ = ~
o~ ~
o°"
~ ~== =~.~ .-~ ~
0
" ~ ~'~.~ ~a ,.=: ~
~a ~ =.'.~ ~ ~-~..~ " ~ .= o . = ~
"~-~ ~c~ , ' ~ 0 ..~..~ ...-~ ra 0 0
oo o ~ ~
'~.~ ,. o=~.,= ~ .
• ==~ ~=,-~~.=.~ ~ - ~
•-~ ~ ~ ..= ..= ~ o.~
0 .= =

ca ~ .,-~ ~.~

o.

c~
0

co ~) ~ ~.~ ~ 0 ~ ~

,.~ ~.~ ,-~ 0 co

0
cO •"~ .=
0
Z~
O~ ..= ~ .-~
0
z-a-~ ~ ~ ~.~ ~ ~ ~) ~°~
e~
o
~ ~ 0 ~'~ ~ :.~'~'~

Q;
-

-~ ~.-=.~.~-~ o,=~ =Q ~o- oo

R
,.o

=
0
0 "~'=

~ -~ ='~ co 0 0
=_,= ~'~'~
-=~ -='-= o
Q., • >
.~

o ~ , ~ ~ ~ ~.~.
=o,,_~ ~.=
~_~ 0 ~.~
<~

21
 0~
P~
e~
•~ ~.~
QJ 0 -~

q;

~
~'~ 8 ~ o =

=o =o~
o=~ .~- ~ =.=-~
0"~ 0 c'~ -- Qj (1) 0~

~.- ~ ~
0 0"0
~'~ ~
I:~ "c~ ~ ~'~ "~o ~
oZ 0 ..~

.~ o~.~
0 "~
Z~N e ~ .~_~ .,.~
0
,.0 0 r~ 0 Q) ~ o,1 N^~x~o

~iI 0 Q; ~ 0 Q3 Qj ~ ~ bll e~

0 ~D ~ 0 ~J ~J 0

22
II! r ~~ r ~i'!i [ - .

' ~

. . . . . . t

. . .~,
0 "

l:

Neg. No. 688 87, 10

Photograph 1 Front facade of numbers 1 and 3Widcombe Parade

Neg N o 689 87,,4

Photograph 2 Widcombe Parade

23
 Neg. No. 689/87/1
Photograph 3 C r a c k on party wall in stairwell

ff ....

b-

,'~eg t~o 689 87 6

Photograph 4 F r a c t u r e in t h e lintel over e n t r a n c e door

24
 N.
4

\
i!!J!~

. i

1
. . . . . . . %

Neg. No. 689 87 6

Photograph 5 Cracking due to deflection of beam spanning shop front

Neg. No, 689 87 3

Photograph 6 Vertical crack at junction of properties

25
 ~-i=.~¢.E~.

~ o
A1
I ~J
il
IIX
II

I1'
i I r-1

I.]'~L~
~°'~'~ ~'P'~"
I I
~---~----- t
I
"\\\
. ii i
, ii

Numbers 1 and 3 WidcombeParade (groundfloor)

26
 LF-. ~ N - - t o

b
-~1 I i , I
i I I

r t I I I i
! _. I I j
I [ i-- ~ t "" t
~ ~ E.xTd,v~J f/Zoo*: ~1o,v~Tst=,,<_To,J

l o , J T,I/~8~ R 18c~. t I
I 1 l I
l l l
I I I I __ I
- -- I .- I I --
t , I I
I J I SlA,KC~L F~o~ iC~b
l I l TO I s l Ft-<x~R H~.~IBF-LN

l I l r ~'~l

t -- r r~tc,:, r~o~'~.bE.b WkU,.S ~ ~ - [

' ! , /

~¢.t (_K ~: LU'E.

]
t F4 ~,LC~V F-

I~ AT~RoOM
I J o FIRST FLOOR F.~I~T5 ]~UE
TG ~LHOV~J~ FOR tt4C_l~F.J~%¢_~>

Numbers 1 and 3 Widcombe Parade (first floor)

27
 c-
o

o.

i 1.

.5 e.l

~_ o

J
o ~

~ _o £ o

II

0
0
m

"0
t-
O

o
v

~_tL t~

° E
~3 Z

f--l__ I ,rl ~!

28
 . ~ = -~ = : ~ o> ~ ~.~ o-~ . ~ > o ~ ~ ~ ~ ~ ~ ~ o

~ ¢~ ~ - , . ~ ~ ¢v o ~ b~ ~ .~ ~ ..~ ~ ' ~ o "~,.~ ~,., :n

~ °~ ~ ~ °~ °~

~ "~ ~ ~ ~ o "~"~ ~ o:~ ~ >~ ~ ~ .~..: ~ ~.~_

~ o ~ ~ ~ ~ . ~ ~ ~ ~'~ ~ ~ o ~o o ~ ' n ~ ' ~ = ~ ~: o

~-" o ¢v ~ ~ ~ ~ ~ . ~ ¢v "" ~-- ~ :n ~ ~ ~ ~ o ~:~

.~ ~. ~ ~ . ~ ,~ ~ oo

o~ ~~ ~ o ~ = o ~ ~ .

~ ~.- ~ o ~ ~ -~ ~t~

~ ~ ~ r~ ~.~ = "-~=

~ . ~,. ~ ~.~
.o ~ ~ .~ ~ ~ ~

29
 oo
oO

!i
"0
L~
~ = ~ o-~

"r~ ~ =,..0 . ~ -.Q 0

,.~u 0
•

~ ~ ~.~..~

0 .~o~
.~ ~-~ .,., .-= E0 8 ~ izi ,-~
g~
0,1

<
o "~1 ~-~ . _,=
o~
. .,-,
~ ==. ='u0~ =~ ~ ¢~ ""~ " ~"~ ~

.~.~o~ ~ ~ -=2 ~ =~-~ .=.=~= -,~°=~

dl
,'-'] 0 ~,.= ~ = ..,-, ,,- 0,~

'10
" ~ ~.~ coO; bO

...~ ,.~ >, o ,= ~ x~ . ~

~ ~ ~o co m =~ ~ O.¢n o~'~ oo ~ ~,.~

~ . ~
o-d ~ ~ "
.--~..~L~ ~ ~

~ ~: ~'~

~.~ .,.~ ~

~ ._
" °3~o m
~..~ ~ w~.~ ~ ~-~ ~.,, ~
"~ ~ ' ~ 0 a~
o~-'~ ~ 0 , ~ 0 ~

o.= o.=~= = =-=~ ~3 ,'~ == =

, . ~ .~.~ ~ ~ 0 ;.~, m ~ ~ 0 ~, . ~ ¢0 ,-~ ~'. bO

0--, "~ ,-~'~ ~u

30
 0

"0

0~ o
~0
~q=l ~ ~-~ ¢~ 0

o
0

~'~"* 0 4-~ ~'~ 0

~'~
bJO
w

"0

o
o
..0
~ . -~ ~
"0
.~ ~-o o S =
o

"0

e~ e~ ~.o

.~ ~ ~ ~ ~"o
z ~ ~ ' ~.~ ~ o
~,..~
•-~ ~ 0
~. o .o .~ ~

.<
.< o]

g~
e~

31
 6 6

0 o

(D

U.
z

32
 Neg. no. R 7 3 7 / 8 7 / 3
Plate 1 East w i n d o w in s o u t h aisle

Neg, no. 737/87/6

Plate 2 W e s t w i n d o w in s o u t h aisle

33
 a

i~ ~ ..... a
I ....... t

k,,4 i
Neg. no. 737/87/8
Plate 3 S o u t h w i n d o w , s e c o n d from east end

lb.

Neg. no. 737/87/5

Plate 4 S o u t h w i n d o w , first east of e n t r a n c e

34
 0 ~"
d=

"'~-~ ~0~ 0 ~ ~ E,^ ~ : . . ~ ' ~ ~ .~

0 ~ ~ ~ ~ ~..~

¢.,
~ 0
~~.~
.-~ t n ~'~

O0
.~5 ~ ~ .~ .~ 0 ~-~ ~..~ 0
~.~ ~=.~-~ ~o.~
z
0 ~ ~ . . c : "~ ~ ,~, ~ ~.~,.~
0 0 ~ ~ N ~D
Z o ~ ~'~'-' ~)"~ 0~ "Q ~ "~
•.~ ~ ~ ,-~ ,~, ~.,.-,
0

r.)
:::1.. ~ ~ ~cl 0.,-~ ~-r~-c~ ~ •

~'- ~ u-=~.
~ ' ~ 'o~' - ~o ~
.~ o ~ . ~ . ~- . .~u o~ ~

0..~ .-
~-~..~n~ ~ 0 ~ 0~.~ ~>
~ ~3 ~.~ ,.~ ~ ~ ~ ~ ' ~,~ ~ ~.~
0
I~ 0 ~-~ - c~ Q 0 ~-.~ ~
0 0 ! t ~ . ~ '~" ~

~o~
~2
"~ "~ ,~

~-~ 0 ~ ~ ~o ~o 0,-~I ~

~oo~,n
""I 0
'~ "el i> - ~ . . Q
"C:I I::I ~ 0 ~

,.Q 0 Q) ~) -

~
~ . ~ o
z
0 ~.~
E~
0
z o .~

.<
E~
E~
0 "~.~ ~
!~ ~ ~ ~.~ o ==~.~
•~ ~ - ~
o

r~

.~ 8 = .- -~ ~ . ~

~.~ ~.~ .o

0 r~ NN
•~ ~1)~ O ~ -~

u ~ ,.~ ~ . ~
m'~.~ ~ c:-.-~

35
 Kitchen (bathroom over)

I !

Dining room
(bedroom &
passage over)
No. 53 Evesham

~orc.~ I
I

3 I
7 Timber frame
Brick wall Lounge Bedroom over)

Timber frame
Evesham R o a d

Fig.1 Plan of Old Bell Cottage

36
 The Old Bell Cottage

Distortion of front elevation

37
 Internal d a m a g e in lounge

V i e w of house in Evesham

38
 0 "~ °~
~ ' ~ ~-~ ~ ~ ~ =
~ ~ ~ D ~ ~o ' ~ 0
~ ~ . ~ m ~-~
0 ~ ~ ~ ~ ~ ~'~'~
.~ ~.~ ~
.~-~ ~ ~ i> m~.) m 0
NR~ e~
~'~ ~ ~
m e'~
• 0
~ ~ ~ o~
P,
~ 0
0

0 ~
~N
~o

~ '0
• ~ ~ ~
,-~ ~ ' 0 ~,~ .~.

~ ~ _ ' ~ ~, ~ ~ ~,~_

~ °~ .~
~R~o~ ~ .~
,-~ 0 0 ~ "~ el - ~ I~,
~.o ~. Z ~ ~.~ •~ ~

o ~ ~ e N
NN_~g 0 o~ 8S
e~ o
•
Q) Q)

.~.~ "Ioo ~,= o o

~.~.) ~ 0 ;~ 0
r.~ u~ u~ u" ~u <~:& L'~ 0 ~ -0.~.~.~ ~ ~ ~ ~,..~

~0

2~2
o

~ u~_ ~ .
.~ o
~'~ ~ o

"~ ~ o e~

Q) • Q~ o
e-~ ,.~ oo
Q) o o'~ ~ ~ ~ ='~

~ g ~ ,.~ "u ~
~.~. ~ ~ ,.t= ~
oN ,.~ 0 ~ ~.¢l ~ -~-~ ~
o,,~ o
,.~ ~ ".,~

0-,~ ~ ~ ~ ~'~
o ,.~ ~ " N ~."~
m ~
0

Q; m.~ mC~ 0"~ ~

~ - - ~ l~ Q) ~

~ ~ 0 0
~ ~ ~--~.. ~
0 ~ m
O'3 Q;
0 t:~

0 ~"

& ,- ~Q
ll::l 0 -,-~ o~

N~N~NN

39
 °~

E-,
~x -I- -~- Jr- -I- -I- I -I-

r~

~o
-I-
~0
-I- + +"~, I I +-t-
+ =;
I~,~ "t-

~r3 II
x~
=o~== I
~ + "
-I- +; T
~ 0
m

+ + -t- + I I

~3

~~ ~
0

~= ~=-~ ~ o
~ ~
0 "~ o=
0
~ .=~.~
0

~.'=~ ~o'~. ~ ~o o~, ~

~ 0 .~ .~. 03
~i~.~ ~,~ ~-~
~ ~ ~..= 0~ ~ ~ o

,.~
~'~ ~ ,~ ~ ~,~ "~.~

¢1
Ne.~ i~ o,-~ ~ = o ~ ~"~ ..= c~= .~ c,

eNe "C. ¢~ . m~C,

R

2 ~eee
•~ m [.-, ...~ ~ E i E o
o ~
• ~...~ •~ ~
~'= ~ E~ ~ " ~,'~ =-, ~ " =
• ~ ~ ;>'~ ~"¢~ 0 ¢~ ;> 0
;'~ ca 0 =-, ~ ca ~0'~

.~--~ <

40
 ~I~-° 2. z ~ • .-.TE

s_/~,~4m~-if,h;.~

/ sT FLOo,~. PLAN

0 I I Ii
t i i o

~t

• ~2- - - . -:.-

~o~//VD FL o o ~ P I. ,,"IN ,',lem,~ f.toN / 7-= /v

~ c . le /" ) o =

41
 .~/'t&~,,,,l" c,'9,,k,
It ,

~ ~ ~

/ I! 1 o.t.,~,.~z,,,. !!~d I

EL E v.4 7~oN A-A m

E'L. E V A 7"l Ot,,l ~-Z~

P o ,,~ c H

J
/
l ~o. 7.

E Z..E V.'¢t 7 / o , * , / C- ~

z~" yoN.

42
/
~/~¢ ,,~,.~ I"I

,~ °"
.~,.. o o "~°

C r =, =.k ;.,

db O ~

EL_ ~ YA T/oN c'~ - Z~

C'i" ~ c ~ .

I._.
r--

~volw

43
 !

r ............. !~ ~i¸¸

Plate 1 Rising Sun f a r m h o u s e

. , ° .., t l t i ~ W

Plate 2 North wall of porch

44
." .~y

Plumb line

• Main road

Plate 3 Showing lean of front porch wall

Plate 4 South wall of porch

Printed in the Unrted Krt~gdom for Her Ma)esty s S t a t i o n e r y Office

3~91 8 9 Dd8222758 3 36 C3 G476 10170