CarRacing Setup: (1) Suspension

Cars are incredibly complicated systems of connected bodies.
In the simulation setup menus you are going to see only a few spring & damper variables to adjust, but these adjustments have to coexist with at least 5 different bodies of weight, which also [should] have some torsional deformation to them.
Before we are going to be able to explain each other how these systems coexist, we need to understand what we are playing with and what characterizes each part of the system.
Then, to make sense these parts coexistence, I have come up with a method of discreet division, into groups consisting of scenarios, relevant variables & targets, which are going to help us imagine the best next step in achieving some new, desired behavior from the car.
When both terminology, characteristics & workflows are established, we'll try to go through a few exemplary scenarios / tests, which are going to establish our knowledge in stone.

Most parts we are going to describe in the next part are responsible for "transfer of energy". Meaning, they take pre-existing grip/performance potential and focus this potential in other component of the car. (Of course, to some finite point of overload)
For the needs of this guide, we are going to consider that the only variables which can "add performance potential" are: tire composite, wheel alignment, aerodynamics, weight (which we cannot fully control)

1. Parts & Terminology

[Mathematical] System

Because we are talking about simulation, computerized / digitalized representation of a physical object, we need to be aware that we are analyzing a collection of mathematical equations representing physical properties of these objects.
Why is this important? Because, to allow for calculations of high frequency, these mathematical representations are limited to significantly simplified imitations. This fact permits us to workout very narrow specifications of individual components and their interactions, freeing us, from infinite complexities of physical interactions of real world objects.

a. Tires & wheels

I have heard car wheels being called "the unsprung mass", so many times that I believe it must have somehow become viral misinformation. If we grab a rubbered wheel & throw it on the ground, it becomes immediately apparent, that the wheels are also sprung. What is more important and fascinating, is how different air pressures impact springiness of said wheel and how important they are to overall performance of the car.
When in reality the tire pressure is going to determine the size of tires contact patch and at the same time, the area of heat exchange with the road, in most simulations we are just going to focus on hitting a pre-determined by the sim creators sweet spot.
Picking the right compound is also going to be a situational thing.

b. Alignment

Parameters of tires & their interaction with pavement determine the maximum potential we can extract from the car.
For front, cornering wheels, we have often the option of configuring: Caster, Toe & Camber
For rear, most often only driven wheels, we are going to have only: Toe & Camber
What's very important about wheel alignment is the understanding, how a flat wheel, which pushed by hand tends to roll straight, is even capable of making the car corner WITHOUT SKIDDING.
An eye opening explanation has come to me through this, recent video by FortNine.

c. Strut

A strut column. The bearer of weight. The structure responsible for interaction between large mass of the car & small mass of the wheel.
There are many different solutions for vehicle suspension, in which struts are most commonly present in form of MacPherson struts. These MacPherson struts are the most popular solution in street cars but, even though their operation differs in important ways from solutions used in high tiers of motorsport, their construction allows to easily divide operation of the whole suspension assembly. This division allows us to easily identify & describe operation of each relevant variable found in racing simulation setups. These descriptions are generic enough & precise at the same time, to make them relevant for many types of suspensions.

d. Spring & Damper

To approach configuration of a spring & damper system, we need to understand how these two specimen of engineering represent can be represented mathematically.
Spring is essentially an energy storage. Left alone, it is not going to be doing much, but each compression results in energy added and force of decompression is going to be equal resultant of the energy stored.
Damper is essentially an energy dissipater / resistor to energy transfer. Out of every unit of energy you put into it, it is going to make part of it disappear [In case of hydraulic dampers, this partial force is going to be reoriented back against the compression force]. The faster you squeeze it / the more energy you put into it, the more energy is going to be "lost".
So, coexistence of these two parts results in a constant battle between a hyperactive killer spring and a calm police enforcing damper.

e. Anti-Roll bar

A torsion (twisting) bar, which under compression from one side, creates a proprtional compression force on the other side, thus transfering spring force from less stressed side, to the more stressed side. Their simple construction and operation on single axle, has profound impact on balance of a performance car. Through interaction with the other axle, through the car body, which ends up becoming another torsion bar, it is going to be our main tool in correcting forces acting on individual tires.

f. Body & weight balance

In reality, cars body also carries characteristics of a spring. Even reinforced chassis of a performance vehicle twist & deforms throughout the time of competition.
As simulations rarely include representation of these deformations & forces, including the body in calculations as a heavy brick, energy transfer through this part becomes much more harsh and impactful than it is in reality. [For example, BeamNG has less issues caused by too stiff setup suspension, because simulation of car body deformation distributes impact energy through itself and act as an additional spring]
This phenomena makes spring & damper settings much more detrimental to drivability, presenting us with even bigger challenge & opportunity, to create an advantage.
A very important aspect of car chassis, often neglected when thinking about setups in simulations, is the distribution of weight.
Most cars are going to have more weight on the front, at least by 5%, but there is a big difference between weight on the front & rear.
Front mass is going to be centered around the engine and its parts, situated very low, close to horizontal axis of the front axle.
Rear mass is going to be distributed more evenly around the higher parts of the car, especially in front-wheel drive cars.

2. Scenarios

In most cases, you are going to be working on a car in which you cannot alter its construction & inherent parameters. You are going to have some limits, some less desirable behavior originating from [hopefully not from mistake of a developer] imitation of old generation designs.
Because of this, it is futile to talk about complete design of vehicle suspension. We are going to be trying to squeeze out performance out of a system in some predefined state, so in this chapter, I am going to propose, how we are going to disassemble what we are given and determine the best course of action.

a. Stability [in suspension travel]

To grade stability of a 4 wheeled vehicles suspension, we need to find out what happens to alignment under compression. For a beginner, it is going to be enough to set the car springs to soft setting, [trying] to do a lap or two, then setting the springs to a much harder setting. All without touching other parameters. If the car has been hard to control with soft suspension and setting it to hard has given it superpowers, it probably means that the alignment changes significantly under compression.
This can be proven if access to telemetry data is available. With bad suspension, on telemetry we can observe significant changes in toe and camber (less important) when cornering.

When working with cars with bad suspension, to extract as much performance as possible, we are going to be forced to limit ourselves to a setup with minimal suspension travel, no matter the track conditions.

b. Trinity of Energy Transfer

While changing suspension settings we are going to be working on energetic balance of the car. Primarily targeting absolute balance, but keeping in mind that all platforms (Front-engine, mid-engine etc.) can gain a bit of an advantage by skewing the balance.

When talking about balance, I mean, primarily in cornering, maximizing available grip from the tires. Straight-line performance is a byproduct of cornering performance, so don't worry about this for now.

When traveling around an obstacle at speed, we can distinguish three phases, in which we can identify discreet sources of imbalance. This being:

• Deceleration - when approaching an obstacle at speed, we are going to need to lose some momentum by braking. To achieve the best result, we are going to keep some of that braking into the corner in what's called trail braking.
• Neutralization - when nearing the obstacle (apex), we are going to start gently transfering load from the front axle, around the outside, to the rear wheels.
• Acceleration - when coming out of the corner around an obstacle, we are going to put significant amount of energy into the rear axle, expecting two rubbers to take their potential and push a very heavy object. In this process we can expect them to only push us forward, but... we would like to accelerate as fast as possible, so being able to turn would also be nice.

Above three are going to be our main focus in suspension setup. But, there are many more scenarios which we can identify and distill into targets of optimisation.
For example:

• Flat ride - reduction of body pitching. Violent pitching might cause overloading of tires and sudden, unpredictable loss of grip.
• Balance on brows (Brow: short, but steep protrusion, causing suspension to compress violently) - jumping flat, or at least nose up, is much safer than nose diving on jumps.
• Loose surface performance - when driving on loose surfaces, turning is mostly done by sliding and powering out of corners. A more oversteery tendencies might be desired, to simplify rotation.

The limit is your knowledge about details of car mechanics and characteristics of targeted stage.

TBD