The LeMans project

During 2005 Creuat and the Racing For Holland team have joined together to develop the Creuat suspension for one of the most sophisticated and fastest cars on racing tracks. 

The LMP1 Dome S101 Judd exceeds the 300km/h in these circuits where this is possible, and the aerodynamic forces more than double the car weight. In these conditions, the suspension is critical to provide the necessary grip and the needed stability. 

The main challenge was to provide a system that exceeds the performance of the conventional coil over suspension, proven for many many years, with enough reliability to run and win the 24h of Le Mans.

Le Mans is a name for tough competition. All big plays have fought to have a winner car in this race since speed and four wheels go together. The perfect show window to compete on reliability of racing technology. During 24 hours all cars have to provide the maximum performance without failure. Winners have to be the best in many areas at a time.

The system installed in the RFH car uses basically the same hardware as in the GT and RallyCross prototypes. Instead of springs and dampers, a passive Hydropneumatic central unit takes care of each suspension mode in an independent manner. So the car can be tuned to avoid most of the compromises dialed into a conventional suspension made of springs and dampers.

The system has to cope with a car weight of about 900kg that becomes over 2200kg when the aero down force is taken in full. This means that all hydraulic components have to be designed for much wider pressures ranges than in a car without such load variations.  

LMP1 suspension vertical stiffness needs to be very high also to hold the ride height under aero forces. At the same time, the configuration of the vehicle departs considerably from other GT cars because of masses distribution and low rotational inertias. Altogether it requires a complete analysis of the vehicle dynamics before the suspension is designed.

When the system was installed, it was taken into account that it should make it easy and fast to switch from/to the conventional suspension. This was necessary to optimize the back-to-back testing sessions that surely were going to take place.

The cylinder movement, although smaller than the coil over can be monitored for wheel movement and data logged for later analysis.

This arrangement proved good on all the tests carried out. 

Hydraulic actuators, this is plain cylinders, were integrated in the pushrods, and the standard coilovers substituted by rigid rods. In this way the cylinders take all the movement of the suspension

To switch to the normal suspension, the rigid rods were substituted with the springs, and the cylinders were blocked with rigid spacers. This allowed for a very quick swap between the systems. 

With this arrangement the antiroll bars were left all time in place, and did not need to be manipulated for the swapping of the system.

The location of the cylinders would note increase, and somehow reduce the height of the center of masses of the entire setup. Most of the original pushrod was used, but rigidly connected to the chassis. Since the wheel movement for this type of car is small, the length of flexible hoses needed was also very small.

Most of the hydraulic circuit was built with small bore rigid pipes to minimize the parasitical elasticity introduced with the conduits. We had to bear in mind the very high elasticity rates in the suspension.

The volume of oil was also minimized to prevent two problems. One was the dependency on temperature. The second one was the elasticity of the oil itself. Both were after all within very much acceptable ranges.

With the CREUAT lower warp mode stiffness, the torsional stiffness of the chassis plays a smaller role in the front/rear weight transfers that take place on corners. The net result is a more consistent understeer. In fact, one of the characteristics of the suspension that immediately have an impact on driver feelings is the steering immunity perceived against road input.

The graphs to the left show the load unbalances logged during our tests at Magnicours. Graphs refer to the measurements during a whole lap, the upper graph with our system in, and the lower with the conventional coilovers/antirollbars. In particular we measured: 

Diagonal unbalance (red)

100 * (( Load FL + Load RR ) / ( Load FR + Load RL ) – 1)

Pitch unbalance (blue)

( Load FR - Load RR ) / ( Load FL - Load RL ) – 1

Roll unbalance (blue)

( Load FR - Load FL ) / ( Load RR - Load RL ) - 1

Diagonal unbalance provides a measure of uneven load distribution among the wheels. Frequency analysis is the best way to analyze variable fluctuations as it renders them for every frequency. are the graphs show a good reduction over all frequencies. 

Measured Roll and Pitch unbalances show considerably lower with our suspension. The better tire load distribution has a big impact on braking and cornering grip that the driver feels immediately. This permitted quicker driving that shortened lap times beyond any expectation regarding chassis changes. 

Diagonal weight balance:

The less diagonal unbalances, the more similar tire loads, and therefore the more grip as all tires are used better and less tire overloads are produced. Tire overload leads to smaller grip indices, so it is important to minimize tire load differences when possible. Diagonal unbalances can be caused by track warp ondulations and by individual bumps. Track warp may be unusual, but not to neglect on banking angle changes. On the other hand, individual bumps, and corner stones equally produce diagonal unbalances that temporarily reduce tire grip

Pitch balance:

Braking and accelerating can drive suspension to its travel limits easily. At the end of the straight, with all aero down force applied to the chassis, pitch movement can easily reach the bump stops, crippling the suspension and leaving all movements left to tire deflection alone. In this situation car handling is very much compromised and altogether reduces driver confidence in some very stressing points in the circuit.

Separating all movements helps greatly avoid reaching bump stops. While pitch movement can be limited to avoid bottoming the front of the car, each wheel is still free on its other three movements (roll, vertical and warp), so bumps and hard braking do not make driver's life even harder. Front wheels can cope with bumps during hard braking without producing the usually high tire overloads as soon as bump stops are reached. 

Roll balance:

Suspension balance is based in the appropriate control of weight transfers produced in front and rear axles during cornering. Weight transfers create some grip loss, and the suspension needs to distribute it between front and rear wheels so the car keeps the necessary grip on each axle. Normally, for rear-driven cars, it is desired to keep more grip in the rear, while front-driven cars want to keep it in the front. In this way the car can maximize the acceleration capabilities on corner exit.

Roll unbalance has to be corrected with the steering. When grip is in the rear axle, the car becomes understeering, and the driver needs to increase the steering angle to cope with it. This is a small payback for the increase in traction that will allow taking corners faster.

Roll unbalance, thus, is a very sensitive issue. Since driver has to "counteract" it, he will have to correct every fluctuation that occur. This is called "wheel fighting", and it creates a big stress on the driver as he continuously has to correct the direction of the car. 

This is probably one of the first advantages felt by the drivers as they immediately recognize aiming the car being more accurate, specially on corner stones and chicanes. 

Determining the spring for the system was relatively easy as the conventional system already had been used for years, and it was clear that on roll balance and other movements we had a good point to start with.

Nevertheless, the roll balance proved an interesting point. Once its fluctuations were reduced, we found that the optimal balance was considerably more neutral than with the conventional suspension. So roll balance fluctuations were keeping the car away from the optimal balance.

Damper rates were less obvious. Conventional coilovers used only four dampers, so the damper rates were obvious a compromise to suit roll, pitch and vertical movements  as good as possible.

To find the right damper rates it was necessary to calculate the damper needs for each movement of the car. Therefore we had to measure the actual inertias of the car for every movement. The left image shows how the roll inertia was measured by hanging the car and letting it oscillate around a longitudinal axle above the car center of masses. Altogether it showed how far was the damping from the optimal values when every movement inertias were taken into account separately.