[0018] The dimensions of the continuous sinusoidal profile, such as wavelength and peak to trough amplitude, are advantageously chosen so as to achieve maximum transmission of tyre vibration into the driver's cab while minimising the exterior noise disturbance. There are a number of approaches which may advantageously be considered when seeking to achieve this. For instance, it is beneficial for the surface to cause significant horizontal vibrations in the vehicle suspension since, unlike vertical vibrations which are generally stifled by the vehicle suspension mechanism, horizontal vibrations are more readily transmitted to the driver's cab. Both the wavelength and the peak to trough amplitude will clearly have a significant effect on the level of horizontal vibrations generated, since these factors will govern the contact-forces between the tyre and the sinusoidal surface.

[0019] Furthermore, the wavelength λ (m) of the sinusoidal surface is directly related to the forcing frequency f (Hz) on the vehicle tyre, and the vehicle speed v (m/s), according to the following relationship: 1$f=\frac{\nu }{\lambda }\mathrm{Hz}$

[0020] The wavelength of the sinusoidal profile is preferably chosen such that the forcing frequency at the tyres of crossing vehicles will excite one, or a number of, resonant frequencies within the vehicle.

[0021] Advantageously, the wavelength of the sinusoidal profile is chosen such that a low forcing frequency will be generated at the tyres. The human ear is considerably less sensitive to low frequency vibrations and, at frequencies of around 35 to 40 Hz, will be 40 dB less sensitive than at 1 kHz where the sensitivity of the ear is approaching a maximum. Therefore, it is advantageous for the forcing frequency to be in the range 35 Hz to 40 Hz so that external noise disturbance is kept to a minimum. Unlike a continuous profile embodying the present invention, a series of humps or bars of similar dimensions to the continuous profile, will produce short duration impulsive forces at the tyres which theory shows can be resolved into a wide range of forcing frequencies. Some of these frequencies will be close to 1 kHz and will therefore be significantly more perceptible to the human ear.

[0022] The following table represents the variation in forcing frequency with wavelength of the sinusoidal profile in accordance with the above equation. The speed of the vehicle is assumed to be 30 mph (48 km/h), or 13.3 m/s, which is the speed limit in many residential areas: 1

[0023] The frequencies generated by profiles 1 to 5 span nearly two orders of magnitude, from those close to so called “body bounce”, frequencies, up to those that would excite tyre cavity resonances. Surface profile 3, having a wavelength of 0.35 m, generates a forcing frequency of 37.0 Hz which would not cause significant perceptible noise at the roadside since the human ear is not sensitive to this frequency. Profiles having a wavelength of around 0.35 metres are therefore particularly beneficial in terms of minimising the external noise disturbance.

[0024] Surface profiles 6, 7 and 8 illustrate the effect on forcing frequency of sinusoidal wavelengths that are slightly longer or shorter than the wavelength of 0.35 m. These dimensions were tested using a number of vehicles of different weights and sizes. Measurements of exterior noise were taken using a Bruel & Kjaer 2144 analyser with a type 4149 microphone at 7.5 m from the vehicle centre line at a height of 1.2 m. Crossing speeds were also measured using a radar speed metre and ranged from 15 mph to 40 mph (24 to 64 km/h). Interior noise and vibration were logged with a similar analyser. Interior noise was measured near the driver's head position and vertical vibration was measured using a type 4366 accelerometer attached to the driver's seat rail. The results of these trials indicated that, in general, surfaces with wavelengths smaller than 0.35 began to produce increases in exterior noise as a result of higher forcing frequencies being generated.

[0025] While the slightly longer wavelengths did not produce a significant increase in exterior noise, they were found to be less effective at creating interior noise and vibration to alert drivers. Generally, surface profile 3 produced the highest levels of interior noise and vibration, without generating significant increases in exterior noise.

[0026] FIG. 1 of the accompanying drawings shows two graphical representations of an experiment to measure interior noise (FIG. 1A) and vibration (FIG. 1B) levels for a mid sized car traversing a traffic calming surface embodying the present invention. The squares represent the noise and vibration measurements taken within the vehicle as it drives over the traffic calming surface, and the dots represent comparative noise measurements when travelling on level ground. It is clear from these figures that the interior noise and vibration levels within a vehicle are significantly increased as a result of the present invention.

[0027] The first two profiles in the following table enable a comparison to be made for a variation in peak to trough amplitude with a wavelength of 0.35 m. Measurements of external noise were made, and recorded, as before and comparative measurements were also made for vehicles traversing a patterned imprinted surface and a rumble strip with a series of ridges. The measurements of exterior noise for each surface profile are shown in FIGS. 2A to 2D of the accompanying drawings. 2

[0028] Referring now to FIGS. 2A to 2D which show a graphical representation measurements of external noise generated in the vicinity of a number of traffic calming surfaces. FIGS. 2A and 2B represent noise measurements taken in the vicinity of two traffic calming surfaces embodying the present invention (profiles 1 and 2). As a control, noise levels were also measured for a vehicle travelling along a level road surface and these measurements are shown by the dots on each of the graphs. FIGS. 2C and 2D represent noise measurements taken in the vicinity of vehicles traversing a patterned imprinted surface (FIG. 2C—profile 3) and a rumble strip with a series of ridges (FIG. 2D-profile 4).

[0029] It can be seem from FIGS. 2A and 2B that there was very little increase in external noise compared to the external noise generated by a level road surface. However, the measurements illustrate the substantial increase in noise alongside the imprint pattern and the conventional rumble strip, especially at the higher speeds.

[0030] FIG. 3 shows an elevational view of a traffic calming surface 10 having a continuous, substantially sinusoidal, profile 11 at the central region thereof. The peak to trough amplitude is denoted by 12 and the wavelength is denoted by 13. The leading and trailing edges 14 and 15 respectively, have been laid such they provide a gentle incline for traffic to approach the profiled section. The incline preferably has a uniform gradient over its length. In this example, the incline is designed to take the traffic from the existing surface to the height of the peak over a distance of between 3 m and 5 m. The surface is not drawn to scale however typical measurements would be a wavelength of 0.35 m and a peak to trough amplitude of 6.5 mm.