ABS verses No ABS

I did some impromptu testing today with the ABS in my car at an Autocross. There was a hard braking zone that we looped through twice and with the ABS I found I needed to brake approx 120 feet before the apex to enter the turn. The ABS woudl simply pulsate through the entire braking zone. With the ABS disabled, I found the braking distance was closer to 60 feet! This was a huge difference in how fast I could go. It translated into 80 or 90 feet once I added a little more speed. I found that the car was definitely faster without the ABS. It was also harder to drive. The braking had a much more percise feel and after a few runs I could keep it right on the edge of locking. This all sounds great... so whats the downside. One extra ounce too much and once the tires locked, it was all over. By the time I could unlock and reapply, the car was tossing cones into the air. In a DE, it might mean taking a run off. So whats the compromize? My boxster had smaller brakes than my 996 but could stop faster since the ABS was not always on. If the car can stop SO much faster w/o it, why does it go on so early? Wouldn't this behaior translate to the same thing on the street to a 'normal' driver in a panic situtaion? The new ABS in the 02/03 cars would result in a longer braking distance as the ABS is too sensitive? I want my old ABS back! SO for now I am educating myself in the ways of the old... no ABS. Any thoughts?

I agree that the ABS on the 996s is on the sensitive side. Brake modulation is a big part of performance driving. With or without ABS you want to modulate your brakes to get maximum use of it. With ABS you want to modulate it right to the point just before ABS would activate. Adjust your braking point accordingly. Without ABS you want to modulate it right to the point just before any wheel lock up. Adjust your braking point accordingly. The only downside of not using ABS IN THE DRY is that you might flat spot your tires occationally. The upside beside having better lap time without ABS I believe you might use less pads and you can do panic stop a lot quicker if all 4 corners are at a complete lock up. Of course, you lose your tires in a panic stop! Tires are cheaper to replace than body work.

In guess one of the questions is how sensitive the ABS should be? Should it allow a 1/2 second of true wheel lock before it kicks in? This is the way the boxster was. Why should the 996 be so different? Maybe part of it is the pads. I ahve the stock pads on the 996, where the Boxster had Pagid Orange. I plan to change them on the 996 soon.

I know this maybe a reach here but have you thought of upgrading brake system and using slicks on the track. That would negate the use of the "ABS" if lap times are what you're after. Regards. Mike

Race pads and tires will help even with stock brake system. There is nothing wrong with the ABS. It just pulsate too slow for tracks. The Cup cars ABS are calibrated for racing application and I believe they pulsate a lot quicker.

ABS should not take any delay. It should start taking over as soon as a lock up of wheel is detected not before a lock up. Now we just need the ABS to pulsate quicker therefore more pad and rotor contact between pulsation.

I am running Michilen Pilot Cups so I am as close to a slick as I can and still be DOT legal. Even with these tires hot, it still ABS' like crazy. The Boxster weighs the same and never did this (albeit smaller brakes). The cup car uses an older control module so I cannot download that program (dealer looked into it for me) I ordered Pagid RS14's today so we will see if that makes a difference. I guess my big question is if under normal dry conitions the car (properly driven) can stop in half the distance w/o the ABS, than the ABS does not seem to be programmed corerctly. Maybe it is setup for street tires ony?

Ugh oh! From the PCA Tech Site... I've experienced four ABS system failures, in which a beep would first occur and then a warning light would flash on my dash. My dealer checked out the first failure and couldn't find anything in the diagnostics. I therefore ignored the second two warnings while the last failure resulted in a catastrophic, high-speed, head-on crash into a tire wall at the end of the longest straight on our home track, Mosport. Have you heard of any similar ABS failures on the 2002 model Carrera 2? No this is the first one I've heard of, but I was there at that event in the chief instructor's Boxster right behind you when it happened. The street ABS is terrible on the track on all of the modern Porsches, beginning with the 964. The 964 RSR, the 993 RSR, and the 996 Cup cars have racing ABS, which is a completely different animal. Street car ABS is calibrated and timed to pulsate relatively slowly compared to racing ABS, because it needs to handle ice, snow, gravel, etc. Racing ABS is a much higher speed system, and it was so good that it was outlawed in ALMS when the GT3R came out (without it) in early 2000. Back to the street cars, the street ABS was so bad that the first 964 Carrera Cup cars had two defeat switches, one eletrical on/off switch on the dashboard, and the second one built into the brake system itself so that super hard pedal pressure would defeat it entirely. In general, you don't want to be "getting into the ABS" (activating it, feeling it pulsating) on the track, and if you are, then something is wrong -- wrong tires, wrong conditions, wrong speed, wrong line, something. Now you did not report pulsations, and from the dark solid tire marks off the pavement it is clear there was no ABS but plenty of braking. But if the system gave you a warning, then the dealer should be able to tell you what the error code was, it's in there somewhere, and can be retrieved with the Porsche System Tester 2 (very expensive) or possibly with OBD-II Scan from Baum Tools in Florida (less expensive but less capabilities). Joel Reiser - PCA WebSite - 10/3/2002

ABS, a suggestion. Write to PCNA and ask them to have the design engineers look into the following design change. Use the PSM system to determine when to activate, OR NOT, the anti-lock braking system. Anti-lock braking is NOT designed for shorter stopping distances, exactly the opposite as a matter of fact. ABS is designed to reduce the braking capability to the front so you can maintain directional control during severe braking manuvers. Additionally the actual operation of ABS is a huge compromise. First, they currently have no way of knowing, discerning, that the vehicle might be reaching the maximum braking coefficient with the wheels closer to locked, or in some cases actually locked, than the current ABS state of the art design constraints allow. PSM is designed to "know" when the vehicle is not following the desired "track". So, let me "have" the responsibility to modulate, or not, the brakes, unless the PSM "decides" that the vehicle is not following my steering inputs, or lack thereof. Sometimes my life might depend on a quick straight ahead stop and as long as the vehicle doesn't oversteer then leave me to my own resources, PLEASE.

Interesting thoughts. I agree that ABS is a compromise, but it should not give up that much. I don't know much about the ABS in the 996 (time to educate some more) as to too many channels it has and how it responds. In thory a 'next gen' ABS should have sensors in each wheel as well as steering input and yaw or g force sensors. This would allow better processing of the data and a better result. The previous post on failure aside, I think the issue may simply be the pulse speed (the street ABS is designed for rain, snow and ice) and the time delay with a wheelspeed difference before the ABS engages. If 1 wheel having 1/64 of a rotation speed difference triggers the ABS, the results make better sense. On almost straightline braking, could the system be interpeting the rotational speed differences as locking? Could the 'non-oem' tires be effecting the system. The Michelin Pilot Cups are OEM sized and are designed specifically for the car.

I guess we have to adapt to what we have and brake earlier (relative to situation where no ABS is available like my old track car). I would test my braking points the first time I bring my 996 on a particular track. Set early braking points at all braking zone to get an intial feel. I than brake later and later until ABS kicked in and I would find the general point where ABS would kick in and stay there.

When I had my 92 Jeep Cherokee Limited I would remove the ABS pump fuse during the non-winter months when it wasn't so likely I would lose directional control during braking. You might want to find the equivalent fuse in your 996. The problem being you also lose PSM.

Ed, I upgraded the brakes on my 996 to the GT3/TT brakes. I run Pagin Orange pads with Hoosiers on the track but have also run the old Pirelli P Zero C in the past. As part of the brake upgrade, I also installed the GT3 master cylinder and brake booster. With this set-up, I am almost never into the ABS on the track. I can sometimes feel in in very rough, bumpy braking zones but other than that, it does not come on. I can also momentarily chirp or lock the fronts if I apply the brakes too hard initially, which leads me to believe that the system is calibrated just fine for track driving. Karl

On the 996 Turbo I'm pretty happy with the system -- both on and off the track. And I often do allow it to go deep into ABS mode in the dry. I noticed a change when I switched from cast iron rotors to PCCB rotors. The wheels would lock momentarily before unlocking. I prefer it this way. I'm less happy with it in the rain. There it tends to unsettle the car and be a little harder to unlock before turn-in. So in general, I don't really agree. But maybe it is the difference between the Carrera and the Turbo?? By the way, how do you turn off the ABS? In the Turbo and GT2 this is impossible to do. S.

Karl, what parts did you change? Were they from the GT3 cup or the GT3 street? What year is your 996?

At the M-5 Driving Experience provided by BMW for new M-5 drivers, one of the tasks was to do a max stop but turn through a corner while stopping! It was easily accomplished with the ABS since you still have control. Everyone could do it!! Even rank novices who had come as guests and didn't really even care about pushing the envelope! Nobody knocked down any cones!! When ABS is correctly calibrated, it has every appearance of the incipient skid, which is max performance!!! As you said before... Once you lock'em up, there goes the control. However....When I changed the front brakes to a larger size and didn't get the metering recalibrated, I found that ABS is not that great a help.

Ed, I would certainly try Pagids before doing anything too radical and don't judge them until you have a few laps in. The ABS on my 99 C2 will engage way early when I first put in new Oranges. After a few laps everything straightens out and I'm able to brake right at threshold without engaging ABS. Mine is a non-PSM car though so it might be different.

SPOOKY: i just read this post, got my wife's car (a mere '94 BMW 316..) and on my way, avoided a serious crash with a foolish toyota driver who too a left turn without warning.. Although the road was rather wet due to an earlier rain, the ABS helped me to steer the car away to the right without having it to revolve around itself or go straight ahead with wheel-lock and luckily avoided the toyota.. Two things I learnt: 1) i'll always wear the seatbelt coz i wasnt 2) ABS helps in panic situations when your mind is not really programmed to react for the instance that arises, so better have it and learn how to use it for track use.. Wael 996 C2 99

Mine is also non-PSM. It was intersting driving it no psm and no abs. Just like it must have been in the old days. I have grown up with all the computer controled stuff. I ordered the pads and we'll see what a difference it makes. My X74 is going in next monday and it will hopefull help expedite the weight transfer to the front tires and may help. I forgot the boxtsre was lowered and the 996 has not been yet.

Karl S, IMMHO it is highly likely that your ABS has failed for some reason. Waz996, ABS helps in panic stops ONLY if you need to turn or prevent oversteer. Otherwise ABS can easily get you KILLED rather than be of any help.

I've been lurking on this thread for a while, and I've been getting more confused. 1. How do you turn ABS off? As far as I know, there is no switch to turn ABS off. You have to physically disable it. If you disable it in a way it was not meant to be disabled, wouldn't you be risking brake failure? 2. I've never taken my car to an autocross. From what I know and have seen, relative to DE, it is a slow speed event in a much shorter course. As such, braking, driving, turning, etc, have to be done in a much shorter distance. The rhythm of the driving movement therefore has to be significantly accelerated and truncated. Wouldn't ABS be more of a help when, due to the truncated rhythm, you try to turn and threshold brake at the same time? 3. Again, I've never taken my car to an autocross, but isn't threshold braking still threhold braking whether we are talking autocross or DE? The way I understand threshold braking is that 100% of your tires traction is dedicated to braking. There is nothing left for turning. If you turn when you are threshold braking, you end up locking and flatspotting tires since you now take away some of that braking traction asking the tire to turn. The reason I ask is that I can feel my tires starting to slip at the threshold, and ABS doesn't cut in. It cuts in only if I turn without letting up a bit on the brakes to give some traction for turning. 4. I understand that the racing versions of ABS modulate faster and will get you around faster compared to the non-racing ABS, but that means you are depending on the ABS system to get you around curves despite exceeding the cars traction capabilities for simultaneous braking and turning. So, if you can stay within the cars performance envelope, why should ABS make a difference, when it shouldn't go on anyway if you are driving well? 5. Is autocross so different from DE that I am misunderstanding the whole thing? <img src="confused.gif" border="0">

Ed, It will be interesting to see what, if any, impact the suspension will have on the ABS problem. It certainly improves control under heavy braking. What size Sport Cups are you using? I found 225s to be too small in front so I switched to a 265f/285r combination. The Cups are great tires.

[quote]Originally posted by Ed Newman: <strong>Karl, what parts did you change? Were they from the GT3 cup or the GT3 street? What year is your 996?</strong><hr></blockquote> I swapped the calipers, rotors, and pads from the GT3. I don't think there was any difference between the Cup and street car parts other than possibly the pads. I also had to change the front uprights and hubs, as the GT3 brakes don't fit the stock parts. As I mentioned earlier, I also changed the master cylinder and brake booster for a GT3 part. My car is a 1999. For 2002, the GT3 Cup cars have 6 piston front calipers and probably bigger rotors, although I'm not sure. I'd guess they have some additional tuning to the ABS controller and master cylinder as well. Karl

[quote]Originally posted by wwest: <strong>Karl S, IMMHO it is highly likely that your ABS has failed for some reason.</strong><hr></blockquote> Nope, the ABS does function. I can feel it easily in the rain and it does sometimes come on in rough braking zones. With street tires on I can also feel it. It's working fine, it just doesn't intervene as quickly as what Ed and others have described. Karl

I am also a bit confused here. I also agree that you can stop faster without ABS on the track but why should that be? I was under the impression that ABS cycled right when you were going to lock up the brakes, thus thresholding braking would be possible except without the nasty flatspotting if you lock them up (Non ABS). Why is it then that you can stop faster without ABS ?

In 99, the cup parts may have been 330mm liek the GT3, the new 6 piston fronts are 380mm! This autox was atypical. We had a straight where I reached speeds over 85 mph, a hard braking zone to sweeping U turn, followed by a slalom parralel to the straight, a tighter U turn, another run down the same straight and then on to the rest of the course. There was a second straight before a buss stop where I got upwards of 70 mph, then braking on a slight turn. Car felt great ther w/o the ABS, could really feel the tire traction as it started to lock. The problem is that ABS does not work the way threshold braking does... it probably should though. ABS senses the wheel locking and takes a oneway trip into ABS mode. Once in the ABS mode, it pulsates the brakes to lock and unlock them. This results in longer braking distances as you are effectively only braking half the time. Threshold braking is the technique of pressing the brakes until the wheel begins to lock and then backing off slightly to maintain the brake pressure at 99%. Threshold braking IS possible in an ABS vehicle if the ABS does not immediately kick in. If is does, like my 996, you cannot as the computer responds at the same time as you and starts the ABS mode, now you are out of control. Its almost as if the older less advanced systems are better. BTW, I remember a few years back, Acura advertising that the NSX had the most advanced ABS system as it independently monitored and controlled each wheel. The 02/03 996 has 3 channels (rear together) in its Bosch 5.7 system, but does it then pulse ALL the wheels? Anyone know what system the Cup Car uses?

BTW, in the Boxster, the ABS would only come on in the rain, rough pavement or on cold race tires. Never during a DE or autocross

Just guessing, but. Anti-lock braking systems are open loop bang-bang servos. off/on, off/on, off/on, etc. If the computer "predicts" that the wheel is about to lock it releases and then re-applies the brakes. That is almost entirely a mechanical function so we're talking some serious time delay here between the time the electronic "release" signal goes out and the pads actually move away from the rotor. The individual brake fluid pressure bleed valve must open and some of the fluid must flow out of the caliper allowing the brake piston to release the pad. Now, how soon do we re-apply the brakes? And how fast do they re-apply? It very well could be that some engineer sat down at his/her computer and determined all of these mechanical delays, system time constants, for the STANDARD system as shipped and those were the delays programmed into the ABS computer. Now you put an aftermarket modification into the system, say "better" brake calipers that just happen to have more fluid capacity, and the time for the brakes to fully release increases by 20 to 30%, more time than the engineer allowed for in his/her bang-bang servo calculation. Good luck with your aftermarket brake modifications.

How did you so simply disable the ABS ? I'd like to try no ABS and see for myself.

To disable the ABS remove fuse 9 from row B. The dashboard will light up like a christmas tree but I did not notice any problems. I am not sure what effect this would have on a car with PSM. I assume it also leaves a code in your computer so make sure your dealer is cooperative and don't suprise them. P.S. your on you own for any damage or problems with trying this

[quote]Originally posted by Ed Newman: <strong> ABS senses the wheel locking and takes a oneway trip into ABS mode. Once in the ABS mode, it pulsates the brakes to lock and unlock them. This results in longer braking distances as you are effectively only braking half the time. Threshold braking is the technique of pressing the brakes until the wheel begins to lock and then backing off slightly to maintain the brake pressure at 99%. Threshold braking IS possible in an ABS vehicle if the ABS does not immediately kick in. If is does, like my 996, you cannot as the computer responds at the same time as you and starts the ABS mode, now you are out of control. </strong><hr></blockquote> ABS is much more sophisticated than this. It is hardly an on/off switch and it isn't correct to say that the brakes are functionin only half the time. I am able to modulate the brakes around the ABS point if I want pretty easily. One thing a human cannot do is to modulate the brakes on all four wheels seperately. ABS can do this. I don't like the feel of the ABS in the wet but this is because it tends to unsettle the car at a time when I need it to be most stable. In the dry this is less of a problem. I do believe that there is a perception issue here where ABS will give the feeling that it isn't braking as effectively as braking without ABS. But all of the tests show that a human cannot beat an ABS system. I use it on the track and my only regret is that I do not make more use of it at the appropriate times. I have the 996 Turbo and not Carrera so my system is a little different. But I do think what I said still applies. S.

After you pull the fuse for ABS the PMS will be off and you will get the PMS light on your dash.

"Threshold" braking and "hardly" ON/OFF... Humans have "senses" that allow them to perfrom threshold braking. Obviously if braking and traction forces aren't "balanced" all the way around then the value of our senses declines. An IDEAL Anti-lock braking system would "know", compute, the roadbed cohesion coefficient for each tire. It would need to assume that maximum braking capability is being applied, "know" the initial speed and weight of the vehicle, and then by "watching" the rate at which the rotational rate of each wheel declines it could compute the "rolling" traction coefficient. So now the ABS computer has a replacement for your human "sense". Given the inexpensiveness of computing power today I suspect that most systems do just that. But now comes the hard part. How does the computer "modulate", moderate, the braking HP back to threshold level? It can't, modern day technology hasn't reached that stage. But what it can do is an approximation. If, when "I" design the ABS firmware I am given the reaction time of the components, I can certainly get close. What I need to know is, once I apply the electronic signal to the individual wheel's brake pressure "bleed-off" solenoid, how quickly will braking action decline by, say, 10%. I must also know how quickly the re-application occurs once the signal is removed. Once I know all that I can write a program in which an approximation of the correct OFF/ON duty cycle is used to keep the braking HP as close to "threshold" as modern day technology will allow. I suspect that eventually ABS technology will advance to the point wherein we will have positional feedback on the brake piston and pads and "switching" (duty cycle) regulation of that position. Until then we'll just have to live with taht pulsating feeling of OFF/ON braking. Additionally. How can we be sure the designers of today's ABS want to get that "close" to actual threshold braking? It's pretty clear that a good part of their "charter" is to maintain, allow for, directional control of the vehicle during severe, PANIC, braking manuvers. So, just what is the ability of the front wheels to "guide" the vehicle during threshold braking? If I were a designer I might be thinking seriously about releasing the brakes, further away from threshold, as a function of what the PSM/VSC sensor is telling me about how much, how far, the vehicle is deviating from the desired path directionally. So, bottom line, are today's anti-lock braking systems really being designed with closeness to threshold braking as the end goal? And you can bet that the answer to that varies from brand to brand.

ABS is not designed as a replacement for threshold braking, it is designed to allow maneuverability in low traction conditions. This is IMO why it is a problem for dry road conditions. Even if at best you coul threshold brake at 90% efficiency, you shoudl in theory (and practive) do better than an abs system with a .2 second brake .1 second release that would be achieving 66% efficency in optimum dry conditions. I am not very familiar with the ABS system in the 996, but I do not believe it activates per wheel. It may have sensors per wheel but I think it pulsates (on/off) the entire system. I also believe that once it starts it will continue until the brake pedal is released past a certain level. Anyone have more details contrary or supportive?

I believe the 996 ABS, or the 996 C4S ABS to be more precise, does modulate each wheel separately. At one track that I go to, the track has a significant amount of patching and uneveness. This leads to the ABS kicking in pretty significantly at almost all braking zones. I can hear as well as feel different wheels modulating when it does cut in. This occurs even if I'm braking heavily but in a straight line. I've been there often enough now that I've modified what would be the ideal driving line to make up for the uneveness and avoid ABS/PSM altogether.

Almost all modern day ABS have at least three channels, one each for the front and one for both rear wheels. Many cars, possibly the 996 included, now have four channels. So ABS can be applied selectively. And, ABS ONLY comes into play if the computer detects that a wheel/tire is about to "stall". When driving, braking, over one of the white plastic (slippery) strips on the road you may feel only one single actuation of the brake pressure release solenoid.

In theory would independent control require 4 seperate pumps, or at least a complex valving system? I do not recall seeing this on the 996. I have to check under the hood this weekend.

It is possible for the ABS to modulate each wheel independently since PSM can brake only one wheel.

911 Standard equipment also includes a three-channel ABS system (Bosch 5.7) featuring revised electronics. Braking is smoother with virtually no trace of pulsing, ensuring a more sensitive and accurate response C4S, Turbo Four-channel ABS (5.7) and a 10-inch vacuum brake booster

VERY complex valving system, ONE pump. For 3 channel systems, each channel must have two valves, one to release brake pressure(ABS) and one to apply brake pressure (PSM). For 4....