My latest idea-caster angle and toe-in are the reason your car self-centers after a turn, and will keep moving in a straight line if you take your hand off the wheel. However, caster angle and toe-in cause a minute amount of tire scrub, and also resistance-hence, a car with zero caster and toe-in should get better gas mileage. Now that we have drive by wire and even self-driving cars, would it make sense to eliminate these features? the car could achieve a few MPG better -would this make sense?
But caster doesn’t cause any extra rolling resistance or wear ? In fact, it reduces wear by assisting in quickly straightening up, and otherwise keeping the wheels straight ?
You’d have to use more fuel if you constantly wandered … Not sure a computer control could replace this at high speed, the computer might be able to do if limitted to some low speed…
Toe in and camber …
Toe in exists to prevent flex in the system allowing toe out, as toe out causes instability… they flip flop left right… The constant turning from instability would use more fuel… I don’t think having systems work to control instability are as good as simply avoiding the instability in mechanical engineering … It would be able to do it, but at some cost… perhaps huge… Anyway the feedback to the driver would be disconcerting… the noise ? vuzz vuzz vuzz… damm road inducing toe out again :(.
Camber… actually camber reduces rolling resistance, also reduces ability to grip the road , and turn… So its also impossible to adjust camber to save fuel… There’s no benefit in reducing camber, it doesn’t improve economy, and its radical to increase camber to improve economy, as tyres that are breaking apart or failing to grip the road are not economic…
There’s a certain “goldilocks” element to this, its the way steering by wheels works best… despite it being old technology.
A simple, crude drawing of the 2 front wheels of a car and what you mean by “caster angle” and “toe in” would greatly help this thread.
However, if I understand it correctly, the mechanical linkage between the 2 front wheels in a car is deliberately set misaligned so that it drives better on average, using the extremely crude control system in a modern car of bolting both wheels together with a cross bar and then there’s a gear going to a rod going to the steering wheel.
If each wheel was driven with a simple individual servo motor, with a sensor on the steering wheel (and a force feedback motor on the steering wheel), you eliminate a bunch of parts, and you could get the precise control of steering you describe.
You could do one better - you could design the system so if you take your hands off the wheel, the car drives in a perfectly straight line. See, the force feedback motor would apply a small force to “straighten out” the wheel, making the position encoder sensor on the wheel read zero. The car would be fly by wire, and 0 on the position encoder would be interpreted by the steering control as “drive straight”, and the steering control would choose angles on the wheel servos to drive in a straight line, automatically correctly for tilt in the road and wear and everything else.
You’d never need to align a car after changing tires again, either.
The drawback is that these kinds of electrical systems can fail or be incompetently implemented. I think an FPGA system using combinatorial logic is the only feasible way, using even a microcontroller introduces certain failure modes. Vehicle manufacturers don’t agree with me, which is why certain cars are wirelessly hackable and can crash.
Anyways, you replace some big heavy steel components with circuit boards full of ICs that are sensitive to power glitches and loose wires and overheating and all kinds of other faults. This is why actual drive by wire steering systems leave the mechanical linkages in, and the servos are just an assist, which makes it an expensive hybrid system that loses many of the advantages mentioned above.
Like this one for example:
The difference would be microscopic. We’d save more energy by eliminating the entertainment system and the energy consumption used to operate the 1 watt radio.
It ain’t broke … etc., etc…
You might save a tiny bit of fuel, but the occupants would be unhappy.
Take away caster, and the car will be very twitchy. It’s what makes a car want to return to straight ahead. Make caster too strong, and the car will be hard to turn. Go really too far, and the wheel will want to oscillate like a shopping cart with a bent wheel.
Toe also helps with self-stability and self-centering. No toe will give a ride that wanders. Outward toe will make the car wander and pull to one side while the tires are being chewed up. Excess inward toe doesn’t hurt the ride too badly, but it eats tires and can make the steering feel less crisp and responsive.
Camber is more about optimizing the contact patch between tires and road. Go too far either way and the tire edges are being scrubbed to death. Design decision time - you need to tilt the tires a bit so they can maintain good contact with the road in turns and keep good contact while going straight ahead. How much tilt do you need and do you set it for a hypothetical perfectly flat road or a real-world road that has some crown (slope) to allow drainage? And do you tune it to a specific tire profile inflated perfectly, or do you “detune” it to accommodate different brands of tires that may be a bit or a lot under-inflated?