The ABS discussion thread

[Thad E Ginathom]

Originally Posted by gostel

If your car does not have ABS then the safety factoring would consist of limiting the speed of the car and ensuring that tyres are not over-inflated.

If your car does have ABS then the safety factoring would consist of limiting the speed of the car and ensuring that tyres are correctly inflated.

[adnaps]

Originally Posted by Thad E Ginathom

My basic problems with drivers and ABS.

The right thing to do with ABS: Forget that it is there.

Generally true except when one needs to brake in an emergency situation. In an ABS car one needs to remember that you have an ABS and jam the brake all the way and steer around the obstruction. With my years of driving experience on conventional cars, on one occasion when I encountered an emergency stop requirement, I found out it difficult to go all the way and on hindsight realized that I could have done better (on a ABS equipped swift).

[kapilvgupta]

adding a bit to the working of the ABS sytem..
the ABS controller (ECU) receives inputs from wheel speed sensors and vehicle speed. Whenever it realizes that one or more wheel is about to lock (indicated by a sudden drastic drop in wheel speed) it releases the brakes on that particular wheel and re-applies the brake when the wheel speed is proper.

source: Fundamentals of Vehicle Dynamics by Thomas D. Gillespie

100% slip refers to wheel lock and as one can see the max brake performance is in around the 20% slip range (denoted by miu-p). The ABS system releases or applies the brakes to maintain the wheel slip in the 20% range.
As the braking coefficient is lower on wet or low friction surfaces, the tendency of a wheel to lock increase and thus one can feel the ABS kick in much quicker than normal or wet or icy surfaces.

Originally Posted by SirAlec

Little bit of correction. ABS does not apply maximum friction. But its a ecu controlled unit that contracts and retracts the pads in very fast succession, hence preventing it from getting heated (main cause of wheel lockup) all of a sudden and avoiding locking.

Can you please explain as to how the brake pad temperature would be the main cause of wheel lock?? I agree that brake pad temperature does play a role in its friction characteristics but for all normal brake pads (not considering hi-performance carbon fiber pads) the difference in performance is not drastic with change in temperature.

Wheel lock is mainly dependent upon the friction characteristics between the tire and the road surface. For eg. everything else remaining the same a simple change of tires (same wear level/size just different tread profile or compound) can bring upon a significant change in min. braking distance. One of the main governing factors of min. stopping distance is wheel lock and higher the braking force generated before wheel lock lower in the stopping distance.

peace,
~kg

[gostel]

Ahoy folks,
It's nice to visit this thread once again and see some good discussion going on. I just wish to offer some more food for thought.

Let us accept that the coefficient of friction between the tyre and ground is, in general, less than the coefficient of friction between brake pads and disc/drum. So my (theoretical) point was that it is more efficient to generate friction at the brake pads rather than at the tyre footprint. But am I correct in this assumption?

Now please meditate a bit, if you wish, and visualize this: the friction between tyre and ground causes the wheel to rotate (let's forget the engine bit for sake of simplicity); and the friction at the brake pads opposes the wheel rotation; and our car is moving forward without a wheel-lock.
So now, if you are still with me then you should agree that, to prevent a wheel-lock, the friction at brake-pad cannot exceed the friction between tyre and ground. My question is: then how come maximum no-lock braking is more efficient than skid braking? How many kinds of friction are we dealing with here?

This could be a fun game trying to figure out the theoretical aspect of the whole business. Practical approaches could often be misleading as we could easily get muddled up between cause and effect.
Cheers.

[kapilvgupta]

Originally Posted by gostel

Let us accept that the coefficient of friction between the tyre and ground is, in general, less than the coefficient of friction between brake pads and disc/drum. So my (theoretical) point was that it is more efficient to generate friction at the brake pads rather than at the tyre footprint. But am I correct in this assumption?

Now please meditate a bit, if you wish, and visualize this: the friction between tyre and ground causes the wheel to rotate (let's forget the engine bit for sake of simplicity); and the friction at the brake pads opposes the wheel rotation; and our car is moving forward without a wheel-lock.
So now, if you are still with me then you should agree that, to prevent a wheel-lock, the friction at brake-pad cannot exceed the friction between tyre and ground. My question is: then how come maximum no-lock braking is more efficient than skid braking? How many kinds of friction are we dealing with here?

I am not exactly sure what your are trying to say but would like to point out of couple of things..

• The friction between pads/shoes and discs/drums is always not generally higher than that of tire vs road (for a given vehicle), if not the brakes would hardly be effective and I can not think of a scenario otherwise.
• Friction between the road is what makes the wheel rotate (a powered wheel would rotate even when off the ground). This friction (static friction) opposes the motion of the wheel (at the point of contact) and thus making the car move forward as the tire moves in the opposite direction of vehicle movement (at point of contact)..every reaction has an equal and opposite reaction wala funda. This link can explain it better.

Skidding occurs when this static friction at the point of contact is overcome by wheel torque and the tire spins freely. Similarly wheel-lock occurs when when the static friction between the tire/road is overcome while the brakes are being applied....in other words Skidding is a result of acceleration when wheel-lock a result of braking - both occurring beyond the limit of tire/road friction.

Graphical representation of skidding
source: Link

To answer your question about maximum no-lock braking being more efficient than skid braking - skid braking involves kinetic friction and no-lock braking involves static friction. As static friction is always higher than kinetic friction no-lock braking is more efficient than skid braking (though there are some exceptions like ice, gravel etc..)

peace,
~kg

[kapilvgupta]

adding to my previous post on ABS and wheel speeds...

a graph of individual wheel speeds and vehicle speed during ABS operation.

source: Fundamentals of Vehicle Dynamics, Thomas D. Gillespie

The sharp drops in wheel speeds are indicative of the near lock scenario for that particular wheel.

peace,
~kg

[vivekiny2k]

Originally Posted by gostel

Let us accept that the coefficient of friction between the tyre and ground is, in general, less than the coefficient of friction between brake pads and disc/drum. So my (theoretical) point was that it is more efficient to generate friction at the brake pads rather than at the tyre footprint. But am I correct in this assumption?

Now please meditate a bit, if you wish, and visualize this: the friction between tyre and ground causes the wheel to rotate (let's forget the engine bit for sake of simplicity); and the friction at the brake pads opposes the wheel rotation; and our car is moving forward without a wheel-lock.
So now, if you are still with me then you should agree that, to prevent a wheel-lock, the friction at brake-pad cannot exceed the friction between tyre and ground. My question is: then how come maximum no-lock braking is more efficient than skid braking? How many kinds of friction are we dealing with here?

Originally Posted by kapilvgupta

I am not exactly sure what your are trying to say but would like to point out of couple of things..

I agree, first your thinking needs to be streamlined and then you can proceed to the concepts.

1. Yes, the car moves because of the friction between tires and road (static/limiting normally, kinetic if you are doing burnouts). this is in forward direction and the non driving wheel provides some rolling friction in the backward direction.

2. When you are braking, I assume you have cut off the accelerator, if not, you need to take the difference of engine power and braking power at the driving wheels, and the braking power alone at the non driving wheels. this force works backwards when applied at road.

the car will skid if the coefficient of friction (i.e. force, when multiplied by the weight of the car) between the road and the car is less than the force applied by the engine/braking. It applies in braking as well as in accelerating.

so on a nice road, you can accelerate hard, and you can brake hard and the car does not skid. on a wet/snowy/sandy surface, the co-efficient is low. So if you accelerate hard, the wheels spin, same happens when you brake hard.

what ABS does is cut the braking power until the wheel maintains contact with the road again, and then again applies the brake, doing it ~200 times a second. in older days we had to use our judgment to apply brake just enough so we do not skid.

similarly, in cars with traction control, if you accelerate too hard, and the car senses a spinnig wheel, it cuts off the accelerator until you regain control. you will have to turn off traction control if you want to do power slides

I personally don't think temperature has anything to do with the intent of ABS, there are no temperature sensors in there.

EDIT: to summarize, kinetic friction is always less than static friction, hence ABS tries to bring the friction with the road (because that's all that matters, speed "relative to the road" ) to static friction from kinetic friction.

friction between the pad and rotor is controllable to a high value and should always be kinetic (meaning no lock). even when it's kinetic, it's much more than an average car's static friction with road.

one case where pad and rotor have less friction than the surface is locomotives. They run a risk of breaking the brake assembly if applied when running at 100+ kmph. that's why in almost all trains electrical braking is applied first. a generator is run that converts mechanical energy to electrical energy which is simply burnt thru a large ventilated resistor in form of heat.

[gostel]

Thanks kapil and vivek for your inputs. That exactly is the word I was groping for: static friction. You have put it down quite clearly.

What I was trying to say is that at the tyre-ground interface, the static friction is greater than kinetic (sliding) friction. The friction at the brake pads can be varied from zero to a very large value. However, the limiting value is the static friction at the tyre-ground interface, for a no-lock condition. Moreover, both static and kinetic frictions are determined by ground conditions and state of tyre. Wet/sandy roads reduce the coefficients of friction. The ABS system has to perform within these limits. I feel the coefficients would also depend on speeds; but I have no data for that. Maybe some body has such evidence?

So Thad is correct in pointing out that having ABS is no license for ignoring safety norms. So just imagine you are going on the third lane and up comes a lovely right curve that you would love to take at good speed. There is a truck on your left lane. Unfortunately, a wet patch suddenly comes up. If you braked hard before hitting the wet patch, your ABS will save you; but if not, then God will save you.

Drive safe and happy, dear friends.

PS: Vivek, I agree that locomotives use electrical braking. An electricity generator is coupled to the wheel. When generator supply is heavily loaded the wheel experiences a braking effect. This works best at high speeds; because at low speeds the generator voltage drops hence loading effect is reduced and braking must then be continued with mechanical means. Actually, the generator output is often used to charge batteries to conserve energy. One can continue the electrical braking right down to lowest speeds by using another trick.

[Thad E Ginathom]

So just imagine you are going on the third lane and up comes a lovely right curve that you would love to take at good speed. There is a truck on your left lane. Unfortunately, a wet patch suddenly comes up. If you braked hard before hitting the wet patch, your ABS will save you; but if not, then God will save you.

Even wanting to take a blind corner at "good speed" is hardly good driving and you would be aware of that truck. This is an example of where thinking "ABS will handle this" could be so dangerous. It is wrong to break hard on a curve: ABS does not alter that, but may save you in an emergency.

"you", doesn't mean gostel, of course: it means the hypothetical driver that he is talking about.

In one company I worked for, they used to buy the salesmen expensive cars. One guy wrote off a rather splendid sports car. "I didn't know you weren't supposed to brake on a bend," he said. Idiot.

[Ram]

I got my driving license -- the one that I still use, in Bombay in 1981. Goodness, that's nearly 30 years ago! One always looks at oneself as a kid, but Time marches on relentless, and nobody can be blamed.

Since that time, as close friends know, I have owned and driven cars in USA, Canada, Belgium, Netherlands, Luxembourg, Germany, UK, Singapore, Malaysia, Japan and Australia.

The one and only accident in my life (touch wood) was in 2000 on an ABS equipped Buick Regal.

I was on the six-lane Stevens Creek Boulevard (eight lanes at that point) going east in Cupertino, while driving home on rain-wet pavement.

A Camry suddenly came off of Lawrence Expwy. South, cut across seven lanes, into my path and I applied brakes, just as gently and firmly as I had always done on zillions of cars for twenty years in the past.

The ABS increased braking distance on the wet road and an entirely avoidable fender-bender resulted.

I know that the advocates of new-for-the-sake-of-new tech. will want to take a position, sympathetic to ABS, but my experience of decades and well over three hundred thousand miles tells me that, had there been no ABS, there would have been no collision.

[vivekiny2k]

Originally Posted by gostel

One can continue the electrical braking right down to lowest speeds by using another trick.

you mean increasing the stator current by shunting or otherwise?

I feel the coefficients would also depend on speeds; but I have no data for that. Maybe some body has such evidence?

not the coefficient because we are still interested in static friction. but the weight (that needs to be multiplied to get force) will change at high speed because of vibrations. So at a point where the car slightly lifts because of a small bump on road , the friction wil suddenly launch to kinetic value and from there it's a one way street. you can roughly visualize it as reduced coefficient of friction though. auto engineers may have accurate knowledge of this.

Originally Posted by Ram

I know that the advocates of new-for-the-sake-of-new tech. will want to take a position, sympathetic to ABS, but my experience of decades and well over three hundred thousand miles tells me that, had there been no ABS, there would have been no collision.

that makes me think, should there be an override button (one time) for the ABS right at the steering for the times you know you can brake better.

But why do you think you could brake better on that wet road? if you knew how to estimate the optimum braking, the ABS would have never kicked in, right?

[Ram]

Originally Posted by vivekiny2k

if you knew how to estimate the optimum braking, the ABS would have never kicked in, right?

ABS kicks in the moment it detects momentary wheel-lock. And once ABS has kicked in, all is lost. It expects the driver to keep the pedal pressed and
trust the computer program rather than his own instincts.
It certainly outfoxed me in my time of need and defeated my experienced reflexes !

Maybe we are over-estimating the intelligence in a computer program, assigning it expectations bordering on magic and the divine.

Maybe ABS works in average situations, for dumb drivers who slam on the brakes and shut their eyes. Maybe, ABS designers assume, that all humans are dumb and ABS is smart. Maybe a lot of humans in their test cases were.

Optima vary from situation to situation. And extreme situations -- gravel, ice, oil, other obstacles, etc., can happen suddenly -- too suddenly for the ABS designer to have foreseen.

Trained and experienced instincts can save one's life, in ways the software designer never imagined.

[Thad E Ginathom]

had there been no ABS, there would have been no collision.

OK, experience is experience, but I still wonder how you can be so sure. You'd have to replicate the circumstances (which is probably almost impossible) to be certain of that.

[verditer]

I was using a '96 Land Rover Discovery while in USA for some 2 years between 2008-2010.
In a couple of situations when the road was either very wet or icy, when approaching a red light and stopping behind the cars ahead, the ABS kicked in and increased my stopping distance ( I didn't expect this), and I just managed to stop inches away from the car ahead by pumping the brake.
When the ABS kicked in, I heard a very audible kat-kat-kat sound from the wheels, and the brake pedal pulsated. At the time, I did not know this is ABS at work and thought something has gone awfully wrong and removed my foot from brake pedal immediately, then pumped quickly to stop the car asap. Just managed it.

If I had kept my foot on brake pedal as the ABS worked, I'd have hit the car in front of me.

In my 90,000 kms of driving a Zen in India, only once my brakes locked in Mysore road and the car veered and just managed not hitting the median at 80-90kmph when braking hard after suddenly finding a speed hump in front of me. Maybe an ABS might have helped at that time by not getting the wheels to lock but instead go over the hump fast, but if it was not a hump but a car or some other object, with ABS I'd still have hit it.

So ABS doesn't know what object we're trying to avoid by braking hard, so it does its job by trying to keep wheels from locking, but human instinct knows what object is ahead and how best to avoid it.

Personally I prefer my instincts and have the brakes under my control fully, than let ABS take over.... To each his own..

[DWind]

Originally Posted by Ram

ABS kicks in the moment it detects momentary wheel-lock. And once ABS has kicked in, all is lost. It expects the driver to keep the pedal pressed and
trust the computer program rather than his own instincts.

Are we sure of this? I always thought there are 2 things to it.

1. ABS module becomes alive as soon as brake pedal is applied regardless of locking or non locking scenario and monitors continuously till break pedal is released fully.
2. This module calls another when it encounters a near lock situation to redistribute braking force force to other wheels / remove braking force to wheels encountering a lock.

Now based on the above, not all is lost. If release of braking force from pedal is more than locking threshold of the wheel then the 2nd module goes back to sleep and we are at the same braking power as a non-ABS car. No?