Simply put, the suspension attaches the wheels to the car. Its springs serve to hold the chassis up off the ground and allow it to traverse bumps without unduly upsetting the car. The steering wheel and other assorted mechanisms also permit the driver to control the car's direction of travel.
From the racer's perspective, however, the suspension manages the tire's relationship with the track surface. The suspension controls the shape of the tire's contact patch, the vertical loads imposed within each tire's contact patch, and the relative vertical loads imposed on the various tires' contact patches during acceleration, braking, cornering, and the transitions between these states.
The suspension controls the angle of the wheel and tire in relationship to the chassis, and, indirectly, to the track. Camber refers to the wheel and tire's vertical orientation with respect to the chassis' vertical axis, and thus to the track surface, while toe refers to the wheel and tire's orientation with respect to the chassis' longitudinal axis.
On a car with independent suspension, such as GPL's formula cars, each wheel and tire's camber can be adjusted independently. If a wheel has zero static camber, this means that when the car is sitting still, that wheel's vertical centerline will be perfectly parallel to the chassis' vertical axis and perpendicular to track surface.
If the suspension is adjusted so that the wheel leans in toward the chassis at the top, this is known as negative camber. Since camber generates camber thrust as a tire rolls, negative camber will generate a force toward the center of the car as the tire rolls down the track.
As the car corners, its carcass and its tread will distort, causing the inside edge to become more lightly loaded and the outside edge to be more heavily loaded.
The race engineer's task is to find the optimum camber setting so that when cornering at its maximum grip, the tire will be subjected to even loads across its surface, thus optimizing the grip available. See the Camber section of the Tire tire chapter for more details about this.
Note that due to lateral weight transfer under cornering loads, the outside tire is subjected to more vertical force than the inside tire and generates the majority of the available cornering force. Therefore, its camber setting is more critical than the inside tire's camber setting.
On ovals, since the car is only turning left, it is common practice to use positive camber on the left side to optimize grip on all four tires in left turns.
On road courses, typically camber is set to be symmetrical from left to right, although it has become more common in recent years to use asymmetrical camber to maximize the grip on all four tires in the majority of corners or in the most important corners. Too much asymmetrical camber, however can result in instability under braking and/or acceleration on a road course, so it must be used with care.
In independent suspensions such as those used in GPL's formula cars, the suspension is designed so that as the suspension moves up and down, the camber changes in a way intended to help keep the tire's contact patch optimized.
If there were no dynamic camber change in the suspension, as the chassis rolled about its longitudinal axis, the outside tire's relationship with the track, which is typically negatively cambered in the static state, would become more nearly vertical or would actually move past zero camber and become positive, resulting in a loss of efficiency at the contact patch.
However, when the car rolls, the inside wheel droops and the outside wheel moves upward ("bump"). Clever suspension designers use lateral suspension links of unequal length, and carefully located mounting points, to make the wheel's camber change dynamically as the suspension moves up and down.
If the suspension links were of equal length and were parallel, there would be no change in camber from static, but by angling the links so that they converge toward the center of the car, the designer can make the wheel's camber become more negative in bump.
Ordinarily this would also make the wheel's camber become more positive in droop, but by using shorter links on top, the designer can reduce this effect so that the wheel does not see as much positive camber change in droop and the tire's contact patch is not degraded as much as it otherwise would be. This is particularly important at the rear under braking.
GPL does not allow us to change suspension link lengths or mounting points. Unfortunately, we have no data about the camber change curves of the various chassis, so we need to use trial and error methods - including tire temperature readings - to determine the optimum camber setting for a given setup on each chassis at each track.
Toe refers to the wheel and tire's orientation with respect to the chassis' longitudinal axis. If the front of the wheel is closer to the chassis centerline than the rear of the wheel, it's referred to as toe-in. If the front is farther away than the rear, it's referred to as toe-out or negative toe-in.
It's common in real-world race cars to run a slight amount of static toe-in both front and rear. This is partly because real-world suspension bits have a little bit of play in them and as the car begins to move, this slop will allow the front wheels to move back and toe out. Running a bit of static toe-in at the front offsets this.
Mostly, however, race cars run toe-in because it contributes to stability. Toe-out at the rear tends to make the car unstable, because as the car turns and one wheel unloads, the more heavily loaded wheel, if toed out, will tend to steer the car more towards the inside of the turn.
Some drivers do like some toe-out at the front. This will tend to make the car react more crisply to steering inputs, thus sharpening turn-in.
As a race car moves around the circuit, accelerating, braking, and cornering, vertical load is alternately transferred to and transferred away from a given wheel. The car's total weight, plus whatever additional loads are being imposed by the track configuration, is constantly being transferred among the four wheels.
Various factors affect the way this vertical load is distributed at any given moment.
Ride height, because it affects the height of the car's center of gravity, has a significant impact on both lateral and longitudinal weight transfer. The higher the ride height, the more weight is transferred whenever the driver accelerates, corners, or brakes. While rearward weight transfer can actually help acceleration by giving the rear tires more grip, it hurts cornering power and braking power because tires become less efficient as they are more heavily loaded.
Therefore, from the standpoint of weight transfer, in general, the lower we can run the car, the better.
However, the cars in GPL do not generate any aerodynamic download. Modern, high-downforce cars generate very high aerodynamic loads and are set up to compress their springs so they are running on the bump rubbers at high speeds. This is so they can run the chassis close to the ground and generate maximum aerodynamic downforce from the underbody.
GPL's cars, on the other hand, are designed to run on their springs. They are not intended to contact the bump stops except under extreme conditions. Contacting the bump stop dramatically raises the effective spring rate and causes a significant reduction in available grip. It can also dramatically upset the balance of the car.
Therefore, from the standpoint of developing effective setups, we need to run the car as low as possible without contacting the bump stops except under extreme conditions.
You can read much more about this important issue in Spring Rates and Ride Height and How Do I Know when It's Bottoming?
When the car accelerates, weight is transferred away from the front tires and onto the rear tires. When the car goes around the corner, weight is transferred away from the inside tires and onto the outside tires. When the car brakes, weight is transferred away from the rear tires onto the front tires.
Although the springs do not affect the total amount of weight transfer that will occur in any given steady state, spring rates do impact the rate at which this load transfer builds or declines during transitions between two steady states, such as transitioning from accelerating to braking or from cornering to accelerating. Spring rates also affect the amount of load transfer over track surface irregularities.
A softer spring will take a bit longer to compress to a given force than a stiff one, so a stiffly sprung car will tend to react to track surfaces more than a softly sprung car. It will also transition from one steady state to the next more quickly than a softly sprung car. It will feel more nervous, more responsive to steering, throttle, and brake inputs, and more reactive to bumps and other track surface irregularities.
A more stiffly spring car will also theoretically lose some overall grip, because a softly sprung car's greater compliance allows the tires to follow surface irregularities better than a stiffly sprung one. However, the increase in grip from softer springs is offset by the loss of precision and feel, and also by the need to run higher ride heights. The skilled race engineer looks for the best compromise among all these factors.
Remember that the springs perform the very basic function of holding up the car. It's important to have springs which are stiff enough to keep the suspension from bottoming on the bump rubbers and the chassis from bottoming on the track surface.
In general, springs should be chosen so that the effective wheel rate at each wheel is in proportion to the static load that wheel carries. If 60% of the weight of the car is on the rear wheels, each rear wheel should have a wheel rate that is 30% of the total wheel rate of all four wheels, and each front wheel's wheel rate should be 20% of the total.
See Spring Rates and Ride Height for more information about spring rates.
Along with the springs, the anti-roll bars resist the chassis' tendency to roll about its longitudinal axis due to lateral forces generated when cornering. The anti-roll bars, however, are arranged so that they do not resist the wheels' tendency to go up and down together; they resist rolling moments only.
Greater overall roll resistance is beneficial in that it helps reduce the effective camber change which occurs when the chassis rolls. Because it reduces the time it takes to transfer the weight to the outboard tires, greater roll stiffness also tends to make the car more precise and responsive.
Excessive roll resistance, however, can make the car nervous, and can also reduce overall grip by making it more difficult for the car to follow track surface irregularities which require the displacement of only one wheel or the displacement of two diagonally opposed wheels.
Perhaps the most important function of the anti-roll bars is to provide a means for the race engineer to set the overall balance of the car. By adjusting the bars so that more weight is transferred at one end of the car than the other, the car can be adjusted towards more or less oversteer or understeer. See the balance section of the Tire chapter for more details.
When the driver applies the brakes, weight is transferred away from the rear tires and onto the fronts. This means that the front tires can do more of the braking than one would expect from considering only their static loading.
If a tire stops rotating under braking, that tire will lose directional stability. If one or both front tires lock up, the car will lose steering control and will tend to go straight on. If one or both rear tires lock up, the car will lose directional stability and will tend to spin.
Most race cars permit the race engineer or driver to adjust the brake balance forward or aft as desired. To optimize braking capability, the brake balance would theoretically be set so that both front and rear tires lock at exactly the same moment. However, because the loss of directional stability presents the driver with a more serious problem than the loss of steering control, the wise approach is to set the brake balance so that the front brakes will lock slightly before the rears.
Note that the proper brake balance for a given track and car do not carry over to other tracks, other chassis, or even the same chassis with different tires or at a different ride height. If the car has more weight on the front wheels, or grippier tires, or is running at a higher ride height, less weight will be present on the rear wheels, so brake balance must be adjusted forward.
Track conditions may also dictate more forward balance. In corners where the driver must brake while cornering, the car will be destabilized in the corner by the forward weight transfer that occurs under braking, so more forward brake bias is desirable to help the driver maintain control. Also, if the track surface is cambered in the braking area, lateral weight transfer will tend to destabilize the car, so more forward brake bias may be desirable.
Like the springs, the dampers affect weight transfer in transition from one steady state to another. However, the dampers' effect during transient states is more dramatic, because the dampers are designed to resist any motion of the wheels both up (bump) and down (rebound).
Although the dampers do not affect the total amount of weight transfer that will occur in any given steady state, dampers do impact the rate at which this load transfer builds or declines during any transition, including transitioning from accelerating to braking or from cornering to accelerating. Dampers can have a noticeable effect on the way the car react to even relatively small brake, throttle, or steering inputs. Dampers also affect the amount of load transfer over track surface irregularities.
A damper stiff in bump will resist the wheel when it tries to move up over a bump, so a car with more damping in bump will tend to react to track surfaces more than a car with less damping. A damper stiff in rebound will resist the wheel when it tries to droop, such as when the car crests a rise. A stiff rebound setting will also resist the inside wheel's tendency to droop when the car rolls under cornering forces.
A highly damped car will resist rolling due to entering a corner or pitching due to application of power or brakes. It will transition from one steady state to the next more quickly than a car with soft dampers. It will feel more nervous, more responsive to steering, throttle, and brake inputs, and more reactive to bumps and other track surface irregularities.
Stiffer damper settings at one end of the car than the other will tend to reduce that end of the car's grip during transitions. This has an effect similar to that of a stiffer anti-roll bar setting, but the effect is present mostly only during the transition and over bumps; unlike a stiffer anti-roll bar, however, the effect vanishes once the transition has been accomplished, unless there are significant bumps.
A highly damped car will theoretically lose some overall grip, because a softly damped car's greater compliance allows the tires to follow surface irregularities better than a highly damped one. However, the increase in grip from softer dampers is offset by the loss of precision and feel, and may also require higher ride heights. The skilled race engineer looks for the best compromise among all these factors.
The suspension's geometry at each end dictates a point about which the chassis can be considered to roll longitudinally when subjected to the lateral forces induced by cornering. This is a complex topic, a comprehensive discussion of which is beyond the scope of this manual.
In general, if the roll center is relatively high, or close to the CG, the chassis will have less tendency to roll under cornering loads. If the roll center is closer to ground level, the camber change curves will be more favorable, and there will be less jacking effect from cornering loads.
Unfortunately (or maybe fortunately!) GPL does not allow us to adjust the suspension linkage lengths or mounting points, so we can't adjust the roll center height, except to the extent that roll center height is impacted by ride height.
Nor do we have any information about the roll centers of the various cars in GPL, although it would seem that these, along with the cars' camber change curves and CG heights (about which we also have no information), the chassis' weight distribution (about which we have somewhat questionable information), and the chassis overall dimensions (which we know), are the major determining factors in the different handling characteristics.
Experimentation with roll centers, as with the adjustment of camber change curves, will have to wait until the advent of an even more sophisticated racing simulator than GPL. Given what we do have to play with and learn about in GPL, however, this doesn't seem to be a great deprivation!
The steering ratio is the ratio of steering wheel motion to front wheel steering deflection. In formula cars such as GPL's, this is determined by the gearing in the steering rack and by the length of the steering arms on the kingpin.
The number, such as 18:1, is the number of turns it would take of the steering wheel to turn the front wheels all the way around. The front wheels only deflect about 30 degrees, of course, so we'll never be able to turn the steering wheel 18 times.
A lower number, 12:1, means the steering is faster; a small amount of steering input will produce a relatively large amount of movement at the front wheels. A higher number, 20:1, means the steering is slower.
Short, wide cars need slower steering, since their short wheelbase makes them react quickly to relatively small steering inputs anyway. Long, narrow cars need faster steering to offset their increased stability.
The kingpin is the component of the front suspension which carries the wheel hub. Mounting points (ball joints or rod-end bearings) at the top and bottom connect the kingpin to the suspension arms.
The kingpin's mounting points are typically arranged so that the top mounting point is behind the bottom one. As a result, the effective steering pivot axis intersects the ground ahead of the axle centerline. This results in a self-centering effect at the steering wheel known as caster.
Caster in a car is is the same thing as the self-centering effect of castering wheels on a vacuum cleaner or chair. As the car rolls forward, the front wheels try to move themselves to a position so that they will be following the direction of travel.
In a straight line, caster effect adds stability to the car. In a corner, caster will make the steering wheel want to return to center. It is this effect that gives the driver much of the "feel" through the steering. More caster in the geometry will make the steering feel heavier; less will make it feel lighter.
Self-centering effect is also imparted by the tire's aligning torque. If the car is in an understeering state, as the front tires approach their peak slip angle, the self-centering effect from aligning torque is reduced, which causes the steering to "go light". As the front tires go past their peak slip angle and lose grip, the effects of caster are also reduced, resulting in a further reduction in self-centering effect at the steering wheel.
When the car enters an oversteer state, caster tends to make the front wheels follow the direction of travel; that is, the steering wheel will tend to turn in the direction of opposite lock. A car with a lot of caster will actually help the driver to counter a slide, although it won't recover automatically. But the nudge in the direction of opposite lock is a valuable clue that the car is sliding too much and needs attention fast!
As far as I can tell, GPL does not model bump steer, Ackermann effect, or the effects of kingpin inclination. The effects of these steering and suspension characteristics are relatively small in most racing conditions, and I'm sure that GPL's developers felt that the additional CPU power required and the increase in complexity of an already very sophisticated physics model was not justified at the time of the development of GPL.
Ackermann effect refers to steering geometry which makes the inside front wheel deflect more than the outside front wheel as the steering wheel is rotated. On the street, in slow corners, where the front wheels are deflected considerably from straight ahead, this allows the inside wheel to follow a sharper arc, minimizing tire scrub. In fast and medium speed corners, the driver does not turn the steering wheel very much, so there is little impact from Ackermann effect.
Under racing conditions, the tires operate at relatively high slip angles. Because the inside tire is more lightly loaded, and because a lightly loaded tire reaches its peak on the slip angle curve at a lower angle, race cars are often set up with zero Ackermann effect or even some positive Ackermann effect.
Bump steer refers to small changes in toe-in which occur as the suspension moves up and down through its travel. Changing certain aspects of the suspension geometry - the relative lengths of the top and bottom rear trailing arms, for example - can cause the wheel to toe in or out as the wheel moves up and down.
Bump steer in droop is less important than bump steer in jounce, or bump, because the tire is relatively lightly loaded when the suspension is in droop.
Toe-out under bump is generally undesirable because it will make the car darty and unstable. Engineers often set up the car for a small amount of toe-in under bump because this tends to make the car more stable.
Toe-in under bump at the rear is useful for drivers who are trail braking because it will tend to keep the tail stable as the rear suspension unloads under forward weight transfer during braking.
Since GPL does not allow the engineer to specify toe-in under bump at the rear, it may be necessary to use other parameters to dial in slightly more understeer than would be used in a real car. Since GPL doesn't seem to model tire wear in any significant way, running a small amount of additional static toe-in at the rear makes sense and seems to have the desired effect.
The kingpin is the component of the front suspension which carries the wheel hub. Mounting points (ball joints or rod-end bearings) at the top and bottom connect the kingpin to the suspension arms.
The kingpin's mounting points are typically arranged so that the top mounting point is behind the bottom one. This is known as caster angle and results in a self-centering effect at the steering wheel.
The kingpin's mounting points are also typically arranged so that the top mounting point is inboard of the bottom one. This is known as kingpin inclination.
Caster angle and kingpin inclination combine to affect the arc the front wheel follows as it is deflected from center. Both camber changes and vertical displacements occur at both front wheels.
Most race cars' kingpin inclination/caster angle combination result in these effects as the steering wheel is turned:
The downward deflection of the inside wheel causes a weight jacking effect; more load is placed on the inside front wheel and the outside rear wheel, while load on the outside front and inside rear are reduced.
These effects are relatively small under normal racing conditions because the driver does not turn the wheel very much except in slow corners. In slow, sharp corners, the effects are generally beneficial because both the weight jacking effects and the camber changes will assist the car in turning.
However, these effects assume more importance when the car is operating at high slip angles.
When the driver tries to recover from an extreme slide, he or she typically feeds in opposite lock to turn the front of the car toward the outside of the turn and thereby reduce the slip angle of the rear tires.
In GPL, however, the addition of opposite lock during a big slide often aggravates the slide instead of initiating recovery. This is because the front tires are already past the peak of their slip angle curves. As the driver feeds in opposite lock, the front tires' slip angle is reduced and they gain grip, causing the front of the car to grip more and rotating front of the car toward the inside of the corner.
This effect is present in real cars, but it seems exaggerated in GPL. It seems to me that it was easier to recover from big slides in real life race cars than it is in GPL. I am fairly certain that GPL's omission of the weight jacking and camber change effects of kingpin inclination are responsible for this.
When the car is sliding and the driver feeds in opposite lock, the effects of kingpin inclination cause weight transfer to the outside front and inside rear wheel. This causes the rear to grip better and the front to give up some grip, much like softening the rear anti-roll bar and stiffening the front.
Under the same conditions, opposite lock also causes an adverse camber change on the outside front wheel, giving up still more grip at the front.
These suspension geometry effects offset the increase in grip as the front tires' slip angle is reduced. The result is that the car is less prone to spin if the driver applies opposite lock during an extreme slide.
Since GPL doesn't model these effects, many GPL drivers have developed the technique of increasing steering lock when the car is in an extreme slide. In other words, if you're in a right-hand turn, and the tail of the car is sliding to the left, turn the wheel more to the right.
This causes the front tires to go further past their slip angle peak and give up more grip, reducing the slide. It also increases drag and slows down the car, which can be useful under such circumstances. Once the car is back under control, the steering can be returned to a more normal angle.
Because of the adverse impact on recovery from slides, and the resulting contribution to GPL's reputation as "too hard", I would argue that (assuming that I'm right about this omission) the omission of the weight jacking and camber change effects of kingpin inclination is a relatively significant deficiency in GPL's physics model, second only to the omission of an audible signal of suspension bottoming.