ABS System

[Akos Haasz]

Hi all!

Our team decided to implement an ABS system into our 2010 car to test the possibilities of the whole thing. I know, that many teams are engaged in the topic, and I've contacted a few of You.

But could you guys write here about your experiences with such a system, and the problems you had to deal with during the development/configuration?

Thanks in advance!

Akos Haasz
Formula Racing Team of TU Budapest

[Akos Haasz]

Hi all!

Our team decided to implement an ABS system into our 2010 car to test the possibilities of the whole thing. I know, that many teams are engaged in the topic, and I've contacted a few of You.

But could you guys write here about your experiences with such a system, and the problems you had to deal with during the development/configuration?

Thanks in advance!

Akos Haasz
Formula Racing Team of TU Budapest

[Lorenzo Pessa]

Don't forget rule B7.2
You need to lock all the wheels in the brake test.

[Akos Haasz]

yeah, it's like essential for us to be able to switch it off, while the brake system continues to work flawlessly

[TMichaels]

Take fail-safe functions really serious.
Make sure that you are able to "share" wheel speeds by CAN, so that you may also use wheel speeds recorded by the ABS for traction control etc.
Prepare to defend yourself for using ABS in the Design Report. Back in 2007, when we were the only team with ABS, and also in 2008, we had very interesting discussions in design http://fsae.com/groupee_common/emoticons/icon_smile.gif

Regards,

Tobi

[2BWise]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by TMichaels:
Take fail-safe functions really serious.
Make sure that you are able to "share" wheel speeds by CAN, so that you may also use wheel speeds recorded by the ABS for traction control etc.
Prepare to defend yourself for using ABS in the Design Report. Back in 2007, when we were the only team with ABS, and also in 2008, we had very interesting discussions in design http://fsae.com/groupee_common/emoticons/icon_smile.gif

Regards,

Tobi </div></BLOCKQUOTE>

What kind of performance did you get from using the system?

Stopping distance; system on versus driver's best effort?

How long did it take to get a reasonable tune?

[TMichaels]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by 2BWise:
What kind of performance did you get from using the system?

Stopping distance; system on versus driver's best effort?

How long did it take to get a reasonable tune? </div></BLOCKQUOTE>
The stopping distance on a straight braking from 100km/h improved by around 2-3metres with the system on. Sometimes the driver was able to come close, but the ABS reached this distance every time http://fsae.com/groupee_common/emoticons/icon_smile.gif However this is not really relevant on track as we saw the biggest improvements by how fast the maximum pressure was applied. So this can probably compensated for by a well trained driver.
Being constant is also the reason why it improves the average lap time. It works consistently in every corner. A locked wheel at one corner entry usually destroys at least the following two laps.
We compared several endurance runs with and without the system with as constant conditions as possible and found out that the variation between lap times drops heavily while the single laptimes have not dropped as much as expected.
The drivers also needed to learn to brake with the system efficiently which means staying hard on the pedal until reaching the apex.
Furthermore the system is able to independently control the brake pressure at every single wheel which the driver cannot do. This gives great advantages in wet conditions, but we were never able to proof this during an event.
The drivers also commented that the system frees up some brain capacity which they could use for other things like shifting, steering etc. May sound a bit funny, but was definitely a factor.

In general you gain more performance in a direct comparison if your driver is not well trained. It makes untalented drivers, like me, faster in relation to well trained, talented drivers.

Getting a good tune took only about two days as the system which the team is using works model based and you only have to identify the correct parameters. The great thing is: It works as good in dry conditions as in wet conditions without changing a single bit.

Maybe Christopher our suspension guy in 2007 and 2008 and one of the event drivers for autocross and endurance will also comment about this as I was only the electronics guy integrating the system into the car.

Regards,

Tobi

[Zac]

What were you using for a controller and actuator?

[TMichaels]

Continental MK60, used for example in the BMW 3-series E46

[jRob]

Were you guys using wheel deceleration or slip for your control target? Working on a slip based system right now for motorcycles right now, at the "research" phase.

[2BWise]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by TMichaels:
Getting a good tune took only about two days as the system which the team is using works model based and you only have to identify the correct parameters. The great thing is: It works as good in dry conditions as in wet conditions without changing a single bit.

Maybe Christopher our suspension guy in 2007 and 2008 and one of the event drivers for autocross and endurance will also comment about this as I was only the electronics guy integrating the system into the car.

Regards,

Tobi </div></BLOCKQUOTE>

Did you develop your own code or were you able to use the what was developed by Conti and modify the parameters to fit your vehicle?

I ask more out of curiosity (I understand the effort that goes into developing the systems as OEM) and am wondering how you guys are able to adapt the system to your needs. The good race ABS systems on the market all use the same code as developed from the manufacturer, but allowing tuning of specific parameters to suit the specific vehicle. If you don't want to give away any secrets feel free to message me privately.

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Were you guys using wheel deceleration or slip for your control target? Working on a slip based system right now for motorcycles right now, at the "research" phase. </div></BLOCKQUOTE>

I'm understand the assumption that its not a slip controller, but an instability controller. An instability controller will handle varying conditions easier than a slip controller. Tobias, correct me if I'm wrong.

[TMichaels]

Hey guys,
sorry, but as far as I know I am still bound to the NDA which we signed with Conti and given the fact that you, 2BWise, are working for Bosch, which is the main competitor of Conti, I think it is not a good idea at all to answer specific questions.
I can only say that the Continental Race ABS algorithms work a lot different than the ones from Bosch http://fsae.com/groupee_common/emoticons/icon_smile.gif

Regards,

Tobias

[2BWise]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by TMichaels:
Hey guys,
sorry, but as far as I know I am still bound to the NDA which we signed with Conti and given the fact that you, 2BWise, are working for Bosch, which is the main competitor of Conti, I think it is not a good idea at all to answer specific questions.
I can only say that the Continental Race ABS algorithms work a lot different than the ones from Bosch http://fsae.com/groupee_common/emoticons/icon_smile.gif

Regards,

Tobias </div></BLOCKQUOTE>

I was curious as to how you made the connection between the hardware and software, but that obviously answers that. IMO, the toughest part of implementing a production ABS unit is the correlation of the hardware to the software.
In all honesty I hadn't even realized that at the moment that it was done in relation with Conti, but I completely understand. Good work getting the system working to a competitive advantage.

[TMichaels]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by 2BWise:
I was curious as to how you made the connection between the hardware and software, but that obviously answers that. </div></BLOCKQUOTE>
Yes, we did it in close cooperation with Conti, so we did not hack a stock unit.

Regards,

Tobias