Projects

Example Reports

NOISE AND VIBRATION SURVEY ...

Scope of Report

To present the results of a 24-hour noise and vibration survey and to discuss the results with regards to residential development on the site.

1. INTRODUCTION

1.1 A 'green-field' site in is being considered for residential development.

1.2 The site is adjacent to a railway. Civil Engineering Dynamics were instructed to perform a 24-hour noise and vibration survey in order to provide information regarding what noise and vibration control measures, if any, may be required.

1.3 This report concerns noise and vibration from the railway line only.

2. THE SITE

2.1 The site comprises three fields to the west of ... in ...

2.2 ... Lane bounds the north end of the site. This is a minor road with low levels of traffic.

2.3 A section of the mainline from London to ... bounds the western end of the site. The railway line has two sets of tracks and is used by local and express trains.

2.4 The site slopes gently down towards the railway. The railway is in a cutting that is estimated to be approximately 5m deep.

2.5 Figure 1 shows the western section of a plan of the site, showing a possible house layout.

3. MEASUREMENT PROCEDURE

3.1 Measurement locations are shown in Figure 1. M1, M2, and M3 are approximately 20m from the site boundary. M1.1 and M2.1 are approximately 40m from the site boundary. The railway is estimated to be approximately 10m from the site boundary.

3.2 Noise

3.2.1 A Bruel & Kjaer 2236 Type 1 sound level meter was used to log 15-minute values of LAeq, LA90, and LA10, for a 24-hour period starting at 15:00 9th May 2001 at location M2.

3.2.2 A Bruel & Kjaer 2260 Investigator Type 1 sound level meter was used to take simultaneous 15-minute samples at the remaining locations to provide information regarding variation in noise level across the site. The 2260 measures all common acoustic parameters.

3.3 Vibration

3.3.1 A NOMIS 5000 Series seismograph was used to log the maximum Peak Particle Velocity (PPV) occurring every half hour for a 24-hour period from at 15:00 9th May 2001 at location M2. The NOMIS uses a tri-axial sensor and reports the peak PPV occurring in any of the three orthogonal directions. The sensor was fixed to the ground using a spike through the centre on the sensor.

3.3.2 Sample time histories of train pass-by’s were taken using a Diagnostic Instruments Di2200 portable FFT analyser and Bruel & Kjaer 4378 accelerometer and 2646 charge converter.

3.3.3 Measurements were made in the vertical direction at all five measurement locations. The accelerometer was attached to a metal plate using magnets, which is then spiked into the ground.

4. RESULTS

4.1 Noise levels are stated in decibels (dB). A change of 3 dB is said to be 'just perceptible', 5 dB - 'clearly noticeable', and 10 dB - 'twice or half as loud'.

4.2 Noise - PPG24 Assessment

4.2.1 The results are interpreted using the guidance of Planning Policy Guidance 24: 'Planning and Noise' (PPG24).

4.2.2 Figure 2 shows the results of the 24-hour noise monitor at M2.

4.2.3 Missing results are due to heavy rainfall, where the microphone was removed for its' own protection. When calculating Noise Exposure Category (NEC) values, noise levels are assumed based on adjacent measurements.

4.2.4 Samples made at M1 and M3 indicate noise levels to be similar along the western boundary of the site. Noise levels are perhaps around 1 - 3 dB lower at M1 than at M2.

4.2.5 The 'Day' and 'Night' noise levels for location M2, as defined by PPG24, are as follows:

  • DAY: (07:00 - 23:00) LAeq,16hour= 54 dB(A)
  • NIGHT: (23:00 - 07:00) LAeq,8hour= 50 dB(A)

4.2.6 Therefore, the western façade of the houses closest to the railway are classified as NEC 'A' for the daytime, and NEC 'B' for the night-time.

4.2.7 PPG24 gives the following advice:

  • NEC 'A' Noise need not be considered as a determining factor in granting planning permission, although the noise level at the high end of the category should not be regarded as a desirable level.
  • NEC 'B' Noise should be taken into account when determining planning applications and, where appropriate, conditions imposed to ensure an adequate level of protection against noise.

4.2.8 Note that the boundary of NEC 'A' and NEC 'B' for the daytime is 55 dB. The boundary for the night-time is 45 dB.

4.2.9 The cutting provides some screening of the noise of the railway. The cutting provides less screening to upper floors of houses closest to the railway. The screening effect of the cutting is assessed using calculations from The Department of Transport's 'Calculation of Road Traffic Noise' and judgements of site topography from our engineer. It is estimated that the cutting is approximately 5m deep and that the trains are approximately 10m distant from the site boundary.

4.2.10 It is estimated that the first floor of the houses could experience noise levels 3 dB higher than those measured at M2.

4.2.11 'The Calculation of Road Traffic Noise' predicts that noise levels would decay by 3 dB for a doubling of distance from the source. On site measurements indicate that in this case, the attenuation is closer to 1 dB.

4.3 Vibration

4.3.1 Results are interpreted using the guidance of BS 6472 'Guide to evaluation of human exposure to vibration in buildings (1 Hz to 80 Hz)'.

4.3.2 Results from the 24-hour vibration monitor at M2 are shown in Figure 3. This shows that the vibration levels from trains is fairly consistent throughout the day, except for one train at 13:22 that is approximately twice as big.

4.3.3 Figures 4(a) and (b) show the results of the vibration samples presented on human response curves from BS 6472. The response curves are generated from human response to continuous sine waves, and are therefore a very severe criteria to apply to intermittent vibration, such as train sources. The figures are included to provide an indication of where the environment on site falls in relation to frequency dependant human response to vibration

4.3.4 Vibration at the foundation of a building is expected to be less than that measured in free-field conditions at the same locations, due to the increased loading on the ground by the mass of the building. Levels can be reduced by anything from 20% to 60%, depending upon the foundations of the building, the soil properties, and the frequency response of the building. It is a very involved procedure to accurately predict the reduction that will be achieved, although 25% is not an unreasonable prediction. The predicted ground floor levels shown in Figure 4(a) are 75% of the free-field measurements.

4.3.5 Vibration magnitudes can be expected to magnify on suspended floors. Vibration on the upper floors of a building are usually said to be 2 - 3 times the vibration measured at the foundation, but could easily be more.

4.3.6 The predicted first floor levels shown in Figure 4(b) are three times the predicted ground floor levels, assuming that the frequencies occurring on the ground occur on the suspended floors. This may not be the case. If lower frequencies occur, where humans are more sensitive to vibration, the train events will be more perceptible and hence give rise to a higher possibility of adverse comment.

4.3.7 Where 'ground floor' is stated, ground bearing floors are intended. If suspended ground floors are used, the values for upper floors should be used.

4.3.8 BS6472 suggests that, during the daytime, curves 2 - 4 represents a suitable environment for a residential development where a low possibility of adverse comment is required. Curve 1.4 is said to be similarly suitable for night-time. Curve 1 is said to be the average threshold of human perception.

4.3.9 Vibration dose values (VDV's) may be used to assess the severity of intermittent vibration. The VDV method takes into account the frequency dependant human response to vibration, and the occurrence of the source.

4.3.10 From continuous vibration records it is possible to determine approximately 113 train pass-by's in a 16 hour daytime period from 07:00 to 23:00, and 19 train pass-by's in an 8 hour night-time period from 23:00 to 07:00.

4.3.11 Example time histories of two train pass-by's are shown in Figure 5.

4.3.12 Sample time histories were processed using HVLab software to calculate the VDV for typical individual train pass-by's. The total VDV's for the whole day of survey, at the ground floor of houses closest to the railway are predicted to be:

  • Daytime (07:00 - 23:00) VDV = 0.05 m/s1.75
  • Night-time (07:00 - 23:00) VDV = 0.03 m/s1.75

4.3.13 Vibration levels on the first floor could be expected to be at least two to three times the values above, if the frequencies occurring on the ground occur on the suspended floors. The VDV method takes account of the frequency content of the vibration signal. Lower frequencies give rise to higher VDV values to reflect the frequency dependent nature of the human response to vibration as shown in Figure 4.

4.3.14 BS 6472 suggests that for residential buildings, VDV's between 0.2 and 0.4 m/s1.75 for the daytime, and above 0.13 m/s1.75 for the night-time correspond to a low probability of adverse comment.

4.3.15 BS 6472 is currently under revision. However, there is no information regarding any changes to the limits available at the present time. The current limits stand as a 'habitable' environment, but are not to be seen as necessarily desirable or 'good' quality.

5. CONCLUSIONS

5.1 Noise

5.1.1 It is important to note that the guidance of PPG24 applies to habitable rooms only. Kitchens, bathrooms, and hallways are not counted. Therefore, some room arrangements can reduce the extent of noise control required.

5.1.2 Note that the barrier heights are based on judgements of site topography from our engineer, and are therefore indicative.

5.1.3 The frontages of the houses adjacent to the railway are classified as being on the boundary between NEC 'A' and 'B' for the daytime, and NEC 'B' for the night-time, as defined by PPG24. The first floors are subject to higher noise levels due to reduced screening of the railway by the cutting.

5.1.4 PPG24 suggests that the daytime noise level, although acceptable, should not be regarded as a desirable level.

5.1.5 The night-time noise levels indicate that noise control measures are required to ensure an acceptable level of noise. If the boundary of NEC 'A' and 'B' is seen as the required night-time noise level (45 dB(A) LAeq,8hours), then at least 5 dB attenuation is required for the ground floor of the houses, and 8 dB for the first floor.

5.1.6 An acoustic barrier to reduce noise levels to the first floor of the houses by approximately 8 dB is expected to be prohibitively high, at approximately 6m. However, a 2m high barrier could provide a 3dB reduction in noise levels in the gardens of the houses closest to the railway, which is a small improvement. A 3m high barrier may provide around 5 dB attenuation, which would be expected to be a clearly noticeable improvement.

5.1.7 Based on WHO guidelines for night-time noise levels and sleep disturbance, habitable rooms on the first floor are require at least 18 dB of noise control. Although this can be achieved through standard glazing, PPG24 requires that 'residents may expect to sleep with their windows open sufficiently to provide adequate ventilation'. A partially open window may be expected to provide around 10 dB. Therefore, if the required noise control is achieved through closed windows, some form of alternative ventilation is required.

5.1.8 The houses beyond those closest to the railway are expected to be sufficiently screened from the railway by other houses to not required noise mitigation measures.

5.1.9 Those rooms and houses that are neither adjacent to the railway, or screened by other houses from it should be treated thus:

Either:

  • Provide the same noise control as houses adjacent to the railway, or,
  • Determine specific requirements on an individual basis using either:
  • a detailed desk study of screening effects of the cutting to specific locations based on a level survey of the site and railway and cutting, or,
  • site measurements of noise levels at specific locations and heights, achieved using scaffolds.

Of the two methods, the site tests are preferred, as the desk study is still based on an approximation to the behaviour of sound fields.

5.1.10 If the desired solution is to re-locate the houses at a distance from the railway where no noise mitigation measures are required, then the required noise criteria should be sought from the local authority and the distance determined in a similar manner to 5.1.9.

5.2 Vibration

5.2.1 Based on a Vibration dose value (VDV) assessment, the houses closest to the railway can be seen as not unacceptable for residential development.

5.2.2 However, it is vital to note that the standard provides for a habitable environment, and not a high quality one. Figures 4(a) and (b) show that train pass-by's are expected to be easily perceptible. Although one may expect many people to find them tolerable, where there is perceptibility there is a possibility of adverse comment.

5.2.3 BS 6472 is currently under revision. However, there is no information regarding any changes to the limits available at the present time. The current limits stand as a 'habitable' environment, but are not to be seen as necessarily desirable or 'good' quality.

5.2.4 BS6472 makes the following point:

'Within residential areas, people exhibit wide variations of vibration tolerance. Specific values are dependent upon social and cultural factors, psychological attitudes and expected degree of intrusion.'

5.2.5 If a higher quality environment is desired then close attention must be paid to the dynamic response of the floors of houses to control the extent to which dynamic magnification of train vibration occurs.

5.2.6 Finite element models can be used to design floors, changing their dynamic response by varying spans, depths and material of beams, for example, but for increased confidence in the design on-site test structures should be constructed.

5.2.7 For even higher quality environments, base isolation should be considered. The dynamic response of the building requires analysis. It is not the case that a standard design of house can be simply mounted on springs without modification. Base isolation must be undertaken with careful analysis as improper tuning of the system can result in a poor performance that provides little improvement, or even a system that exacerbates ground vibration.

5.2.8 One train was measured that had a PPV of over twice those of most trains passing the site. Sight of the trains was occluded by the cutting preventing identification, but this may be a freight train or a particularly fast express train.

5.2.9 Whilst it may be seen as unreasonable to provide mitigation measures commensurate with the vibration levels of one high event, it does provide an indication to the vibration environment that may occur should the use of the railway change in the future to feature a higher proportion of such trains.

5.2.10 It should be considered that railway traffic may either increase or change in type or operation of rolling stock. The weight and speed (and numerous other factors) of the trains using the line affects the frequency content as well as the magnitude of the signals present in a nearby buildings. For example, when compared to standard local trains, freight trains can give rise to not only significantly higher magnitudes of vibration, but also more energy at lower frequencies. This is reflected in higher VDV values, as these take the frequency content into account, corresponding to a higher possibility, or a probability, of adverse comment.