ABS Ice mode prevention

[chuck b]

Do you track guys disengage ABS (pull the fuse)?

I'm a long time autoxer, rare track guy. I've played w/ the MSP in DS for a couple seasons and it is my first car w/ ABS. I haven't got it to the track yet.

I have many brake questions, but for now I'll keep it short to the question above and my observations. Later I want to dig into root cause, ABS algorythms, etc.

Observations: cold brakes aren't as grabby as hot brakes- that's obvious. But w/ this car, holy hannah if the last corner of the autox course and the brakes have gone to hyper sensitive phase and go into ice-mode ABS. Ice-mode ABS provides horrendous braking and I blow the last corner not by a few inches but several yards, enough to ruin the run, scare me and the workers, and often I'll plow over the last set of cones.

Not every course has this happen since it depends on how braking heavy it is. I've tried changing from stock to high performance pads and it seems that is helps a bit but does not solve the problem. I'm thinking of going stock up front and high perf in back to change the bias.

I've heard of cars w/ sensitive ABS, but it was all the time. And I know brakes get more sensitive w/ heat. But this is way beyond just getting a little hotter. It seems this is bigger problem than just pads. I don't think there is a specific problem w/ my car, but who knows.

Do you guys have any clues, solutions? I will keep digging around and post in other forums depending on what I find here.

Thx in advance

Chuck

[MP3racer]

Chuck,

Good Post!

We have tried to develop the ABS for track use. The trouble we found was that to get ABS to work properly on a race course you really need a "race porgram" (as BMW, Ford,and others use for their race cars with ABS)for the ABS computer. Commonly with front wheel drive on the track and autocross courses a rear wheel will be lifted through some corners. While the wheel is lifted if you need to tap the brakes to rotate the car or set up for a corner the lifted wheel immediately locks. Standard ABS senses this as a single wheel lock-up and shuts you out of the brake system for a few milleseconds (but feels like eternity!) to try to alleviate the single wheel lock. This is the last thing you want to happen on a race course!
We tried to get some development on a "race program" for the ABS to prevent problems like this. To date we have not seen anything from Mazda.

To "solve" this problem we went away from ABS. We currently run the system from the lower model LX with a few changes for competition. The biggest problem then became rear lock-up under threshold braking. Varying the compounds has helped with this.

Hope this helps?

Tom

[chuck b]

Thx Tom!

This addresses my initial main question: ABS sucks so you don't use it.
While I try to avoid ABS, when one is frequently at the limit of adhesion it means one will occasionally go over it!

If you decided against ABS early on, you probably don't know the intricate details of how the computer is trying to brake, so I'll probably have to go elsewhere. (Sorry the lengthy remaining portion; i'm simply frustrated as I know other cars aren't this sensitive.)

For example, you mention the wheel lift affect. This hasn't been the entire problem. I've noticed subtle ABS intervention many times with no problems. Yet I've also reached the end of a run (w/ hot brakes) in a straight line, so no wheel lift, and had the ice-mode kick in. There must be two modes and I don't know what the computer is trying to do- I gotta find out. Then I can help to sort out why it's going into ice mode. This gets especially complicated since isn't this is a four channel system w/ electronic brake force distribution?

The effect is so pronounced w/ hot brakes, it's as if the ABS sensor is acting funny like something is expanding and sending an improper signal. Or what if the toothed sensor ring changes dimensions slightly?!?? Or who knows!

Anyway, I'll keep asking around. Maybe some day, I'll take some data acquisition and I see what braking g's were achieved before ABS intervention as another clue.

Chuck

P.S. Tom, do you know if during ABS pulsing of one wheel, are there pressure spikes in the fluid pressure of the other wheels or are the wheels totally independent?

out

[MP3racer]

Thx Tom!

This addresses my initial main question: ABS sucks so you don't use it.
While I try to avoid ABS, when one is frequently at the limit of adhesion it means one will occasionally go over it!

If you decided against ABS early on, you probably don't know the intricate details of how the computer is trying to brake, so I'll probably have to go elsewhere. (Sorry the lengthy remaining portion; i'm simply frustrated as I know other cars aren't this sensitive.)

For example, you mention the wheel lift affect. This hasn't been the entire problem. I've noticed subtle ABS intervention many times with no problems. Yet I've also reached the end of a run (w/ hot brakes) in a straight line, so no wheel lift, and had the ice-mode kick in. There must be two modes and I don't know what the computer is trying to do- I gotta find out. Then I can help to sort out why it's going into ice mode. This gets especially complicated since isn't this is a four channel system w/ electronic brake force distribution?

The effect is so pronounced w/ hot brakes, it's as if the ABS sensor is acting funny like something is expanding and sending an improper signal. Or what if the toothed sensor ring changes dimensions slightly?!?? Or who knows!

Anyway, I'll keep asking around. Maybe some day, I'll take some data acquisition and I see what braking g's were achieved before ABS intervention as another clue.

Chuck

P.S. Tom, do you know if during ABS pulsing of one wheel, are there pressure spikes in the fluid pressure of the other wheels or are the wheels totally independent?

out

Chuck,

We tried to work on the program and dig into it as you are. Unfortunately, none of the engineers in the US had any info or knowledge of the system. We then tried to work with the engineers in Japan but the language barrier was too great and we didn't get any further. Therefore, I don't have any more data or insight to give you regarding the program of the ABS ECU.

BTW, what manufacturer and compound brake pads are you using? Sometimes the problems can be solved with different compounds.

One thing to remember, be careful when thinking about disarming the ABS system. The normal proportioning valve on the firewall that decreases the force to the rear wheels (since there is less weight on the rear and therefore prevents lockup) on a standard system is gutted on the ABS system since the ABS ECU controls the front to rear proportioning. Therefore, when you shut off the ABS you will have no proportioning front to rear and can lock-up the rear brakes very easily! This is why when we got rid of the ABS system we went to the standard LX system since it had a proportioning valve to help the rear.

Let me know if you have further questions or come up with more data!

[chuck b]

thx for the warning on pulling the fuse for the ABS. I've only done so for one rallyx and the brakes worked perfect for that situation. i never tried it w/ ABS to see the difference.

regarding the ABS program, i suppose you've already tried the good folks at Mazdaspeed Motorsports for input?

the pads I was running were stock. i switched to Carbotech Bobcats. perhaps a slight improvement.

If it is due solely to the pad, I could see using really crappy pads that never develop much bite. The other extreme of using race pads that never get hot enough to really be to grippy i've thought of too but for a daily driver, it doesn't make much sense. it might be good as a test case however. since each course is different, to really be confident, i'd need to switch out pads during the lunch break! and maybe changing pad combination front/rear for bias might help too.

i still need to sort this out. it could be w/ the improved grip of the tires once they heat up at the end of the run that i get increased brake force, thus lightening the rear end enough to then trigger the ABS. But it doesn't feel like that- it feels like the ABS intervenes into ice-mode once hot, at a lower brake force. i have to buy some data acq equipment!

chuck

out

[MP3racer]

thx for the warning on pulling the fuse for the ABS. I've only done so for one rallyx and the brakes worked perfect for that situation. i never tried it w/ ABS to see the difference.

regarding the ABS program, i suppose you've already tried the good folks at Mazdaspeed Motorsports for input?

the pads I was running were stock. i switched to Carbotech Bobcats. perhaps a slight improvement.

If it is due solely to the pad, I could see using really crappy pads that never develop much bite. The other extreme of using race pads that never get hot enough to really be to grippy i've thought of too but for a daily driver, it doesn't make much sense. it might be good as a test case however. since each course is different, to really be confident, i'd need to switch out pads during the lunch break! and maybe changing pad combination front/rear for bias might help too.

i still need to sort this out. it could be w/ the improved grip of the tires once they heat up at the end of the run that i get increased brake force, thus lightening the rear end enough to then trigger the ABS. But it doesn't feel like that- it feels like the ABS intervenes into ice-mode once hot, at a lower brake force. i have to buy some data acq equipment!

chuck

out

At a rallyx, most of the surfaces are slippery enough (gravel, dirt, etc) that you might not be able to tell that the rear brakes were locking. I would give it a try with ABS and see how it goes...

Yes, we have worked closely (and continue to do so) with many of the great people at Mazdaspeed Motorsports including one of their main US engineers. Unfortunately, the ABS programming is done in Japan so little info is available here in the US.

Hawk brake engineers were working with a few teams a couple of years back to try to trick the ABS system by running different compuonds front and rear. They did make some headway but they found it was very track specific. That is, a different compuond combination for each track. This would be very expensive for a "privateer team".

We do run different compuonds front and rear on the race car currently. For the rears we run a very low friction coefficient to help prevent the lockup. This might help your situation??

The more likely situation that you are seeing is from the difference in your tires coefficient of friction or grip. By the end of the run they will be up to near optimum heat range and have optimum grip. Your brake pedal applictation will need to change accordingly otherwise the ABS will do it for you, or lock you out for a bit to prevent wheel lock up. The finess involved in the application of the cars pedals is often the difference in FTD or 2nd place!

Tom

[cougar10agREDUX]

i thought proteges used a 3 channel system?... have you tried replacing all your ABS sensors and cleaning the signal gears? are you using stock OEM axels?