Engineering FYP proposals

[Greg Locock]

Actually the main reason the bicycle model is used as much as it might be in development is that it relies on the user turning components into linear-axle-coefficients. For example, I might use the rebound bumper to induce terminal oversteer in a leaf spring vehicle. Put that in a matrix!

[tmapv]

The so-called bicycle model (better, single track model) can easily take into account lateral load transfers, roll steer, roll camber and many other setup aspects. Moreover, it can deal with high lateral accelerations. The only real limitation is the open differential. Unfortunately, the single track model is often presented without a clear explanation of the underlying assumptions. In my book "The Science of Vehicle Dynamics" I spent a lot of pages to state what the bicycle model is all about.

I presume in order to take the roll into account you have to analyse the model from the Y-Z plane point of view (looking straight at the car, form the front of the car). Would this mean that I have to introduce suspension components into this?

[guiggiani]

Yes, you have to take into account some suspension parameters:
- suspension roll stiffness;
- no-roll center height;
along with:
- track width;
- center of mass height.

See chapter 6 for more details. Errata for the book is available here http://www.dimnp.unipi.it/guiggiani-m/SVDErrata.pdf.

Interactive figures, including those dealing with lateral load transfers, are available here http://www.dimnp.unipi.it/guiggiani-m/interactive.html

[Tim.Wright]

I presume in order to take the roll into account you have to analyse the model from the Y-Z plane point of view (looking straight at the car, form the front of the car). Would this mean that I have to introduce suspension components into this?

Roll does have an effect but there's a million more important things to take care of. You need to get your baseline cornering stiffness of the front and rear axle (i.e. the complete axle not just the tyres) right from the linear range up until the limit (because cornering stiffness changes as a function of the axle/tyre slip angle).

Basically, your job as a vehicle dynamics engineer is to manage the front and rear cornering compliances (in deg/g) of each axle by using the tyre choice, inflation pressure, roll steer, camber gain, compliance steer and load transfer effects.

The bicycle model, as simple as it is, can be used to include roll effects even without having the physical roll degree of freedom. E.g. basically you take a lateral tyre curve and modify it to take into account suspension kinematics and compliance and put it in a bicycle model. I.e. Your 2 tyres for your rear axle might give you 2500N/deg of cornering stiffness. If you add roll steer using bump toe-in this will increase the effective cornering stiffness of the entire axle to (for example) 2800N/deg. Lateral compliance steer might bring you up to 2900N/deg and so on.

The big restriction of the bicycle model it that it assumes that all of these effects happen in phase with each other when in reality you roll steer effects arrive with the same delay as the roll angle. Lateral compliance steer happens first on the front axle then on the rear. However, the phasing of these things are to be looked at in a second step. Get the steady state correct first - then worry about the transients.

[tmapv]

I presume in order to take the roll into account you have to analyse the model from the Y-Z plane point of view (looking straight at the car, form the front of the car). Would this mean that I have to introduce suspension components into this?

Roll does have an effect but there's a million more important things to take care of. You need to get your baseline cornering stiffness of the front and rear axle (i.e. the complete axle not just the tyres) right from the linear range up until the limit (because cornering stiffness changes as a function of the axle/tyre slip angle).

Basically, your job as a vehicle dynamics engineer is to manage the front and rear cornering compliances (in deg/g) of each axle by using the tyre choice, inflation pressure, roll steer, camber gain, compliance steer and load transfer effects.

The bicycle model, as simple as it is, can be used to include roll effects even without having the physical roll degree of freedom. E.g. basically you take a lateral tyre curve and modify it to take into account suspension kinematics and compliance and put it in a bicycle model. I.e. Your 2 tyres for your rear axle might give you 2500N/deg of cornering stiffness. If you add roll steer using bump toe-in this will increase the effective cornering stiffness of the entire axle to (for example) 2800N/deg. Lateral compliance steer might bring you up to 2900N/deg and so on.

The big restriction of the bicycle model it that it assumes that all of these effects happen in phase with each other when in reality you roll steer effects arrive with the same delay as the roll angle. Lateral compliance steer happens first on the front axle then on the rear. However, the phasing of these things are to be looked at in a second step. Get the steady state correct first - then worry about the transients.

Reading through Millikens book, I have a better understanding of steady state. I briefly went over transient and also pair analysis. Transient chapter goes over a lot SMD stuff, and brings in suspension. Is this basically a more advanced analysis of the handling and makes the use of tests such as the step steer test? Then going over the pair analysis (basically advanced stead-state?) then introduces tyre data, which I presume requires tyre calculation separate and prior to the main model (use of TTC?). are both of these something I should go over as well and include in the project, or is it not too big of a step from the standard steady state?