In car braking systems the pedal movement from depressed to pressed corresponds to a very small (<1mm) pad movement. The pad is just touching when the brake when the pedal is not pressed, and pressing the pedal forces the pad and caliper together, increasing friction. Depressing the pedal does not withdraw the pad.

With the brake cylinder fully compressed (as they did for the test), it requires several full pumps of the pedal to bring the pad into contact with the disk. This increases braking distance, which is generally considered a bad thing.

The travel from a single pedal press is small, but its probably a solvable problem. It just needs a different design.

If you make the pedal more sensitive (increase the resulting travel per pedal press) you'll make it harder to apply a precise amount of braking power which doesn't sound like a nice feature. I suppose you could have a more clever system that would move the brake faster until it touches the disc and then behave normally but I suppose that for such a critical system constructors and regulators are very conservative.

There could also be beneficial side effects to having the pad in contact with the disc at all times, such as preventing water/fluids/dirt to coat it. I'm just guessing here.

> If you make the pedal more sensitive (increase the resulting travel per pedal press) you'll make it harder to apply a precise amount of braking power which doesn't sound like a nice feature.

If you've tried hopping in a car from the past few years after driving a much older car you (and you passengers) will have probably noticed this! It's very common in modern cars to have much shorter brake pedal travel, so what would bring you to a gentle stop in an older car will be somewhat more abrupt in a modern car.

This is achieved by having a shorter travel pedal, which is then heavily assisted by the brake servos so it is not horribly heavy to apply. This is nominally for safety's sake, as it reduces the time between seeing a obstacle and the brakes being fully applied. And since all recent cars have anti-lock braking and stability systems it doesn't matter so much if the driver can't control the braking pressure quite as precisely.

> I suppose you could have a more clever system that would move the brake faster until it touches the disc and then behave normally but I suppose that for such a critical system constructors and regulators are very conservative.

Exactly, as I mentioned to the other commenter, it's preferable to have a simple, non powered mechanical system in place to stop the car in the event of bad things happening. Adding extra bits to this system, with the vast R+D cost involved, for marginal increases in fuel economy would be hard to justify for most car manufacturers (especially since the fuel economy saving would benefit the users pockets, not theirs).

It's an interesting one certainly, it sort of comes down to mechanics at the end of the day. In the event of a catastrophic, complete electrical/engine failure, the driver needs to be able to apply enough braking force to stop the car in a sensible distance using their leg muscles. This places an limit on the travel distance of the pads for a full press of the pedal.

Indeed, most cars use a servo that uses vacuum 'pressure' harvested from the engine to assist with braking and reduce the amount of effort the driver needs to expend. However, you can still brake with a non-functioning servo, all be it with a much reduced braking performance.

It's a similar reason why even very modern cars (bar a few, the Infiniti Q50 being one) still use direct mechanical linkages (heavily assisted in many cases) for steering, despite the steering rack, column and rods taking up room and complicating engine bay arrangements. If it all goes really wrong, you need to be able to steer and stop the car.

I think they have direct mechanical linkages because the law says they must. Otherwise many manufacturers would gladly forego all safety issues for profit. A clear public win via govt regulation.

The primary reason is that keeping the pads just in contact with the rotor prevents water from collecting on the rotor surface. Wet rotors can affect braking performance and safety.

As long as the system can quickly return them gently to the rotor surface...

If you change your own brake pads you know that after you finish it can take two whole presses of the brake pedal to get the pads to contact the rotors

The brake pads and jaws get contaminated with dust and the like,making them stick a little. My guess is they get disassembled before test and cleaned perfectly.

How much benefit is it, wonder if having a small compressed air squirter - like a headlight washer - would pay off (in terms of fuel savings) or not.

Would be interesting to know what designs of pads/jaws dissipate the contaminants better?