GB2309784A - Rolling road vehicle vibration testing - Google Patents

Abstract

Apparatus for measuring vibrations in a wheeled vehicle 20 comprising at least one roller 10,12 having a contact surface 18 thereon, formed from four shells 16 for contacting a wheel 22 of the vehicle. The computer (14, Fig 2) which analyses signals from the sensors 24, 26 controls the drive means for driving the roller to vary the speed of the roller gradually over said test period about a nominal speed. The vibrations induced by the repeating nature of the contact surface therefore change in frequency over the test period and produce less interference with the vibrations of interest which are produce by the contact surface irregularities. This enables narrow band data processing techniques such as road noise path analysis to be carried out on a rolling road.

Description

Rolling Road Vehicle Testing The present invention relates to testing of vehicles on rolling roads.

In order to test the amount of noise and vibration in a vehicle produced by the interaction between its tyres and the road when it is moving, it is known to attach microphones and accelerometers to the vehicle and, while it is running, analyse the signals from them to test the amount of noise and vibration at various frequencies. This can be done on a real road or track. Alternatively, to make the tests more repeatable, they can be carried out indoors on a rolling road. This generally comprises two or more large rollers on which the wheels of the vehicle are supported, and which are rotated to simulate movement of the vehicle over a surface. Rolling roads have the advantage that the test can be carried out without the vehicle engine running, thus eliminating a source of noise and vibration which can interfere with track tests. Wind noise can also be eliminated.

The rollers generally have shells attached to their outer surface which is textured to simulate a real road.

It is a characteristic of this type of rolling road that, when the rollers are rotating in contact with the vehicle wheels, the repeating pattern of the surface of the shells produces non-random vibrations at one or more fundamental frequencies and harmonics thereof. For example, if the surface of the shells includes a feature which corresponds to a stone on an otherwise relatively smooth road surface, the repeated contact between the vehicle wheels and that feature will cause a rhythmic vibration in the vehicle. These vibrations are superimposed on the essentially random vibrations produced by the roughness on the roller surfaces, and create a difference between the vibration spectrum on the rolling road and that which would be obtained on a continuous real road surface. If the object of. the test is simply to obtain a total or average measurement of noise or vibration over a large range of frequencies, or to measure vibrations during acceleration conditions, the interference from the roller shells does not affect the result very significantly. However, if vibrations at one particular vehicle speed are being investigated, and noise and vibration in narrow bands of frequencies is being measured, or a plot of amplitude against frequency is being made, the interference can seriously affect the measurements. This is because high amplitude vibrations at fixed frequencies, caused by the pseudo-random nature of the vibrations, produce spikes in the vibration spectrum.

Accordingly the present invention provides apparatus for determining the nature of vibrations which will occur in a wheeled vehicle when it is travelling at a nominal speed, the apparatus comprising a roller having a contact surface thereon for contacting a wheel of the vehicle, drive means for driving the roller when in contact with the wheel, sensor means for sensing vibrations associated with the vehicle, and analysing means for analysing the frequencies of vibrations sensed over a test period, wherein the control means is arranged to control the drive means to vary the speed of the roller about said nominal speed over said test period.

This produces a variation in the frequency of the nonrandom vibrations induced by the repetition of the roller surface, whilst the general energy level remains fairly constant. When the data for the whole test period is averaged or otherwise combined, this variation avoids the production of spikes in the vibration spectrum. This enables narrow band data processing techniques such as road noise path analysis to be carried out on a rolling road.

The variation in roller speed is preferably gradual and preferably small compared with the absolute value of the nominal speed. This ensures that the induced vibrations approximate closely to those which would be induced if the test were carried out at the nominal constant speed.

The vibrations can be in a part of the vehicle, sensed by an accelerometer, or sound vibrations in the passenger compartment of the vehicle, sensed by a microphone.

The analysis can comprise measuring the amplitude of vibrations in certain narrow frequency bands, or producing a spectrum of amplitude as a function of frequency.

The present invention further provides a method of determining the nature of vibrations which will occur in a wheeled vehicle when it is travelling at a nominal speed, the method comprising the steps of placing a wheel of the vehicle in contact with a roller, driving the roller at a speed which varies about said nominal speed over a test period, sensing vibrations in the vehicle over said test period, and analysing the frequencies of the vibrations sensed.

A preferred embodiment of the invention will now be described by way of example only with reference to the accompanying drawings in which: Figure 1 shows a vehicle on a rolling road testing apparatus according to the invention; Figure 2 shows the control system for the apparatus of Figure 1; Figure 3 is a plot of the amplitude of measured vibrations as a function of frequency for a track test; Figure 4 is a plot of the amplitude of measured vibrations as a function of frequency for rolling road tests using constant and varying roller speeds.

Referring to Figures 1 and 2 a rolling road test apparatus comprises first and second rollers 10, 12 which are supported so that they can be rotated about their respective axes by a drive mechanism 13 which is controlled by a computer 14. Each of the rollers has four shells 16 attached to it, the outer surfaces 18 of which are textured to simulate a real road surface. The shells are all the same size and the joints 17 between them are regularly spaced around the rollers at 900 intervals.

A vehicle 20 is supported on the rollers 10, 12 such that its wheels 22 are in contact with the textured surfaces 18. The vehicle 20 is supported so that it remains stationary and its wheels 22 caused to rotate by driving the rollers 10, 12. An accelerometer 24 is attached to the vehicle 20 to measure vibration of the vehicle produced by the interaction between the wheels 22 and the rollers 10, 12. A microphone 26 is mounted in the passenger compartment of the vehicle 20 to measure the amount of noise produced.

The accelerometer 24 and microphone 26 are connected to the computer 14 so that it can analyse the analogue signals received from them.

A sensor 28 is mounted next to one of the rollers 10 which can sense when a marker 30 on the roller 10 passes it. The sensor 28 is connected to the computer 14 so that the computer can determine when the marker 30 and sensor 28 are aligned.

The procedure for carrying out an example of a test will now be described. The test is designed to analyse the frequencies of vibrations would be produced in a vehicle at a nominal test speed of 30 m.p.h.

The computer 14 is programmed to control the drive mechanism 13 for the rollers 10, 12 so that they simulate a road speed which starts at 28.5 m.p.h. at the beginning of a test period and accelerate gradually and continuously to 31.5 m.p.h. by the end of the test period. This gives an average speed over the test period equal to the nominal test speed of 30 m.p.h.

The computer is also programmed to analyse the vibrations by sampling the signals from the accelerometer at a regular sampling rate of 2048 times per second over a time frame of 2 seconds, to store the acquired data for that time frame, and then to wait for the next time the marker 30 on the roller 10 passes the sensor 28. At that point the computer 14 repeats the sampling process for another 2s time frame. This process is repeated for the whole of the test period which includes 200 time frames.

For each frame the computer analyses the sampling data and calculates the amplitude of vibrations as a function of frequency over a given frequency range, for example 20 to 400 Hz. The data from all of the frames is then combined to produce a plot of the average amplitude of vibrations as a function of frequency over the frequency range. The plot is printed out on a printer 32.

Since the time frames are started by the sensor 28, each sampling frame starts with the wheels 22 at the same point on the rollers 10, 12. This helps to improve the repeatability of the test.

Since the speed of the rollers is set to change gradually over the full test period, every time frame is unique and the average speed of the rollers 10, 12 for each frame is slightly different. Therefore, although for each frame there will be spikes in the data at frequencies produced by the non-random repetition of the roller surfaces, when the data from all the frames are combined the effect of these spikes will be smoothed out and the analysis of amplitude against frequency for the vehicle will resemble that produced from a track test. However, because the speed of the rollers 10, 12 is only varied slightly about the nominal test speed, the vibrations produced by the random roughness of the roller surfaces are very similar to those which would be produced in a test in which the speed was constant at the nominal speed.

Figures 3 and 4 illustrate the effect of varying the roller speed on the data obtained. Figure 3 shows a plot of amplitude against frequency for a test vehicle driven at a test speed of 30 m.p.h. on a track. In Figure 4, the broken line shows a similar plot obtained using a rolling road operated at constant speed of 30 m.p.h. It will be noticed that, in the rolling road test, there are regular peaks in the spectrum at approximately 7 Hz intervals. These are caused by the repetition of the features on the rolling road surface. The track test produces a smoother plot, but there is a higher level of background noise due to the fact that the engine is running and the vehicle is outside.

The solid line on Figure 4 shows a similar plot obtained on the same rolling road, but with the speed of the rollers set to increase gradually over the test period from 28.5 to 31.5 m.p.h. It will be seen that the plot is much smoother than the standard rolling road plot, whilst having the same low level of background noise.

Claims (13)

1. Apparatus for determining the nature of vibrations which will occur in a wheeled vehicle when it is travelling at a nominal speed, the apparatus comprising a roller having a contact surface thereon for contacting a wheel of the vehicle, drive means for driving the roller when in contact with the wheel, sensor means for sensing vibrations associated with the vehicle, and analysing means for analysing the frequencies of vibrations sensed over a test period, wherein the control means is arranged to control the drive means to vary the speed of the roller about said nominal speed over said test period.

2. Apparatus according to claim 1 wherein the control means is arranged to vary the speed of the roller continuously over substantially the whole of said test period.

3. Apparatus according to claim 1 or claim 2 wherein the control means is arranged to vary the speed of the roller at a constant rate.

4. Apparatus according to any foregoing claim wherein the control means is arranged to acquire data relating to vibrations sensed in each of a plurality of time frames during the test period, and then to combine the data from said time frames to analyse the frequencies of the vibrations.

5. Apparatus according to any foregoing claim wherein the control means is arranged to calculate the average amplitude of vibrations in one or more frequency ranges over the test period.

6. Apparatus according to any foregoing claim wherein the sensor means includes an accelerometer which senses vibrations of a part of the vehicle.

7. Apparatus according to any foregoing claim wherein the sensor means includes a microphone which senses sound vibrations within the passenger compartment of the vehicle.

8. A method of determining the nature of vibrations which will occur in a wheeled vehicle when it is travelling at a nominal speed, the method comprising the steps of placing a wheel of the vehicle in contact with a roller, driving the roller at a speed which varies about said nominal speed over a test period, sensing vibrations in the vehicle over said test period, and analysing the frequencies of the vibrations sensed.

9. A method according to claim 8 wherein the speed of the roller is varied continuously over substantially the whole of said testing period.

10. A method according to claim 8 or claim 9 wherein the speed of the roller is varied at a constant rate.

11. A method according to any one of claims 8 to 10 wherein data relating to vibrations sensed in each of a plurality of time frames is acquired during the test period, and then the data from said time frames is combined to analyse the frequencies of the vibrations.

12. Apparatus for measuring vibrations in a wheeled vehicle substantially as hereinbefore described with reference to the accompanying drawings.

13. A method for measuring vibrations in a wheeled vehicle substantially as hereinbefore described with reference to the accompanying drawings.