In October, researchers at Cambridge University will begin a project that aims to develop improved techniques for modelling, and ultimately reducing, vehicle cabin noise.
One of the chief causes of this is vibrations passing from the tyres to the suspension and then into the vehicle’s structure. Traditionally, engineers have tried to reduce the vibrations by improving tyre or suspension design.
But Will Graham, who is leading the project at Cambridge’s engineering department, feels that these efforts have repeatedly overlooked the high-frequency vibrations generated through the wheel hubs. He believes that addressing this problem is the key to reducing cabin noise.
Graham’s team has received just over a quarter of million pounds of EPSRC funding to model the high-frequency vibration generated through the hubs.
‘Car manufacturers are more interested in the dynamically unsteady response of the tyre and car at low frequencies from a handling point of view. But that frequency is from 10Hz–100Hz at most,’ said Graham. ‘When you are talking noise it is in the range of 100Hz up to 1kHz. That is the region that hasn’t received successful modelling by car makers.’
Graham added that the complexity of the problem, coupled with the understandable priority given to safety issues, has meant that the quality of the acoustic environment is a relatively recent concern.
To predict the tyre vibrations and hub forces caused by road contact, the team will be developing two computer programs. One will calculate the tyre and hub force responses to a general impulse, with the other working out the specific contact forces and response as the tyre rolls on various road surfaces.
Vibration calculations can usually be carried out using finite element analysis (FEA). But as frequency goes up, the computational demands on FEA also increase significantly.
‘The structure we are looking at and the frequency range taken together preclude FEA,’ said Graham. He explained that one starting point will be to improve the representation of a tyre’s treadblocks by developing dynamic models of treadblock behaviour. For this the team will carry out lab-based experiments to characterise the response of the treadblock’s reaction with the road.
‘The idea is that from those simulations you can extract a simpler model that can be used in the overall computation of road tyre suspension. And this is what we are going to have to compute to estimate the cabin noise,’ said Graham.
While there has been some research into predicting hub forces, Graham said that it has typically focused on low-frequency vibration.
‘The difference here is that we are working from higher frequencies such as 500Hz and looking to extend that model to lower frequencies. What that gives us I suppose is a more sophisticated tyre model than has been previously used for this type of work.’
The team will be working alongside industrial partners including Jaguar and Goodyear. They will provide statistics that the team will able to compare with their own measurements to assess accuracy.
The results of the project are likely to benefit the two manufacturers, as increased consumer demand for better noise vibration and harshness (NVH) characteristics will invariably lead to greater integration of tyre and car manufacturers at the design stage.
‘From their point of view, Jaguar want to get better at predicting cabin noise for a given design,’ said Graham. ‘They don’t want to get to the stage where they have built a prototype and discover there is a noise problem. So having a computational design tool that runs quickly and accurately enough to be useful at these frequencies would be a major addition to their armoury.’