65 replies to this topic

[MatsNorway]

Lee nicole touched upon this in another thread so i felt like giving him some fuel for his fire if you know what i mean.

I wonder how common this is.

http://koenigsegg.co...gring-accident/

[saudoso]

If an ABS fault is triggered in the normal course of driving, the customer would normally take heed of the warning light, stop the car and have the fault attended to.

LOL, most likely drive for another month hoping it would just go away instead of forking the 5K it would cost to fix in a car like that.

 

If it was a NY cab it would stay there forever. They flaunt dash warning as if those were medals.

Edited by saudoso, 20 July 2016 - 23:47.

[gruntguru]

"a back-up feature where the rear wheels are allowed to continue rotating in the event of an ABS fault that results in the front wheels locking up"

Sounds like normal (non-ABS) brakes plus anti-lock on the rear. Should be at least as good as non-ABS, the only issue being the driver was expecting something better.

 

I still have the habit of modulating when I feel wheels starting to lock but need to maneuver (even with ABS). I usually shake my head and tell myself "that's right - ABS - I don't need to modulate".

 

I guess a Koenigsegg driver circulating at the 'ring is in the habit of stamping on the brakes and turning without modulating, making it difficult to adjust to the ABS failure.

Edited by gruntguru, 21 July 2016 - 05:30.

[Greg Locock]

Abs teaches you bad conventional braking skills. There again TC teaches bad launch technique and ESC needs different limit handling approach.

[Fat Boy]

I've been involved in 2 write-offs due to ABS malfunctions. One of them I never got an answer to. The other was an ABS fault in that the main power was shut off while the car was at speed and that disabled the ABS. The actual hydraulics of the car meant that it had way too much rear brake bias with the ABS inactive, so there was locking, smoke and a flip.

 

It was a real bummer when I figured out I was having trouble hearing the driver on the radio because the roof antenna was buried in gravel.

[Ross Stonefeld]

Until that last line I just assumed you were street racing with phantom II 

[Greg Locock]

I do not understand why people do not fit balanced braking systems any more. It means that the abs intervenes sooner,and if it fails you are in trouble.

[imaginesix]

Sounds like normal (non-ABS) brakes plus anti-lock on the rear. Should be at least as good as non-ABS, the only issue being the driver was expecting something better.

 

I still have the habit of modulating when I feel wheels starting to lock but need to maneuver (even with ABS). I usually shake my head and tell myself "that's right - ABS - I don't need to modulate".

 

I guess a Koenigsegg driver circulating at the 'ring is in the habit of stamping on the brakes and turning without modulating, making it difficult to adjust to the ABS failure.

The article indicates the ABS hadn't been invoked for several corners after it failed, meaning the driver obviously wasn't just mashing the pedal. The first time the ABS was called to duty after the failure, it caused a front wheel to lock up and an immediate, unrecoverable spin.

[Allan Lupton]

Abs teaches you bad conventional braking skills. There again TC teaches bad launch technique and ESC needs different limit handling approach.

Yes we who learned to drive a long time ago still remember how to cope with more braking/power/cornering than adhesion, and worry that the basic driving skills which could come back when the automation fails are terra incognita to most drivers these days.

Should someone who depends on automation to turn the lamps on when it's dark and the wipers on when it's raining really be licensed to drive at all!!?

[gruntguru]

The article indicates the ABS hadn't been invoked for several corners after it failed, meaning the driver obviously wasn't just mashing the pedal. The first time the ABS was called to duty after the failure, it caused a front wheel to lock up and an immediate, unrecoverable spin.

The bit I read said the car went straight ahead with both fronts locked.

 

A front wheel lockup does not invoke a spin.

[thegforcemaybewithyou]

From the Koenigsegg analysis:

 

"Our ABS system, like most, includes a back-up feature where the rear wheels are allowed to continue rotating in the event of an ABS fault that results in the front wheels locking up. Letting the rear wheels rotate instead of locking up together with the front wheels prevents the car from rotating. Instead, the car will continue in a straight line. The system worked to specification, as can be seen by the straight skid marks left by the front tires on the track prior to the car colliding with the fence."

 

What exactly does that mean? Was the brake pressure (force) for the rear axle set to zero (or to a very low value to prevent the rear locking) and the car lost about 30 to 40% of the total braking capability because of this?

[Greg Locock]

The report specifically states that the skid marks were straight. I get the impression it spun in the air.

[MatsNorway]

So did i. Flipped due to hitting the rails.

 

It was locking up due to ABS failure during braking, it was not just the driver freezing up.

What kind of shitty design/failsafe locks the wheels like that? Usually you have double lines for the front wheels as well. Very slim chance of both systems failing at the same time.

[GreenMachine]

I am confused, or maybe just a bit behind the times.  I understood that if there was an electronic fault in the ABS, the ABS would become inactive and allow the brakes to function as before, but without the ABS function and any f/r proportioning of braking force provided by the ABS system.  That is consistent with a front lock up and the rears not locking if the non-ABS braking distribution is front biased.
 
I read the report as saying it was effectively driver error, the driver did not modulate the brakes either because he did freeze, or he didn't have enough time before the impact (improbable I would have thought as the car decelerated from 170 to 110kmh prior to the impact).  It would be understandable if he did freeze, for a moment anyway, given the unexpected situation he found himself in.  He may also have chosen to keep them locked since he apparently did not know what the nature of the problem was. However the report is vague about this aspect, and that is further compounded by talking about letting the rear wheels rotate - implying to me that there is something actively intervening in the rear brake circuit to limit brake pressure when the ABS fails and/or the ABS fails and the fronts lock.
 

It was locking up due to ABS failure during braking, it was not just the driver freezing up.
What kind of shitty design/failsafe locks the wheels like that? Usually you have double lines for the front wheels as well. Very slim chance of both systems failing at the same time.

Not sure how you came to this conclusion Mats, I did not read that into the report. The ABS did fail, but I can't see anything that says that the failure itself caused the wheels to lock. In the absence of that link, I choose to think that it was either driver error or driver choice that resulted in the locking fronts.

[saudoso]

Can I ask what's the point?

 

Blown tyre leads to crash.

Leaking brake line leads to crash. 

Smashed windshield leads to crash.

Blown differential leads to crash.

Broken steering leads to crash.

 

There's a whole lot of things that once broken will probably bring a crash. But this item seen as electronic driver aid are worth a thread?

[Ross Stonefeld]

Those things are critical the normal vehicle operation, ABS is a driver aid? ABS/TC/etc failure should be inconvenient or even scary, but not 'fatal'. In either the literal or driving experience sense. 

 

But the proximate cause here is Supercar + Amateur + Racetrack, at least to me. 

[saudoso]

The chat is about cars in racetracks so I guess yes, driver aids, not safety features. 

[MatsNorway]

Edited by MatsNorway, 22 July 2016 - 17:54.

[imaginesix]

The bit I read said the car went straight ahead with both fronts locked.
 
A front wheel lockup does not invoke a spin.

sorry, right. No spin

[Ross Stonefeld]

So they lost it and jumped on the brakes? Or they just grenade the pedal and let the ECU and/or God sort out the rest as standard?

[MatsNorway]

Looking at the skidmarks and the report what do you think?

[chunder27]

It is one of the most notoriously difficult parts of the Ring, I think there might be all sorts of aspects behind a crash such as this.

 

But it is fascinating that Segg have allowed the general public some degree of depth into their findings.

 

A truly refreshing approach.

Edited by chunder27, 22 July 2016 - 18:21.

[Ross Stonefeld]

Looking at the skidmarks and the report what do you think?

 

The report doesn't seem to say, and the skidmarks raise questions hence the questions...

[Greg Locock]

Before ABS the brakes on production cars were usually sized/valved so that the fronts locked first. If the balance is too rearward then sloppy trail braking technique, or even heavy straight line braking will spin the car. The FWD Elan in proto form is an example of this. After I crashed the shiny new prototype they fixed it.

K state that they deliberately replicated this behavior when the abs fails. It is correct, but not sufficient

Edited by Greg Locock, 23 July 2016 - 12:18.

[J. Edlund]

So did i. Flipped due to hitting the rails.

 

It was locking up due to ABS failure during braking, it was not just the driver freezing up.

What kind of shitty design/failsafe locks the wheels like that? Usually you have double lines for the front wheels as well. Very slim chance of both systems failing at the same time.

 

You have misunderstood their report.

 

The ABS system was turned off due to a sensor malfunction; this is really how it is supposed to work as you don't want the ABS system working based on what can be incorrect input signals. Now, the driver will be able to lock the wheels if he brakes too hard. As the ABS system was turned off, a warning light appeared on the dash, but the driver did not notice this light and continued to drive around the race track at high speed. Then, when he came to the first corner where the ABS system is normally active during braking he locked up the front brakes simply because he braked too hard. The rear brakes did not lock up due to the failsafe (probably a forward biased braking system or a conventional brake force limiting valve on the rear brakes) which prevented the car from spinning.

[Lee Nicolle]

ABS, and most elctronic 'aids' should not be on a race car. The driver is supposed to drive the car, not the electronics.

Though it seems this is a road car doing laps of a race track. The ABS failed and seemingly put all the braking on the front.

When and if ABS fails the brakes should function normally. These it appears are programmed not too, so FAIL.

 

Mats may have been referring to my comments about race ABS failing on Porsches the Amaroo accident where the car had a rock hard pedal and no retardation,, or the one that locked all 4 wheels and rotated immediatly in front of me at Mallala.

 

I nrecent months I have had the ABS come on in both of the main vehicles I drive, one was a sensor fouledwith CV grease as the boot had failed and the other was simply mud after [just] negotiating a very muddy road.

Both fairly common occurences, especially mud.

Though both cars [Landcruiser and a Falcon tray top] operated normally with the light on. 

[gruntguru]

I think the most likely (and safest) scenario is:

 - Front wheel sensor fails

 - ABS to front wheels disabled

 - ABS function retained on rear wheels

 

Simply reverting to non-ABS braking does not guarantee against a rear wheel lockup - especially with the pedal "mashed" to the floor. Eliminating rear brakes completely would dramatically increase stopping distance and in my mind constitute negligence.

[kikiturbo2]

Mats may have been referring to my comments about race ABS failing on Porsches the Amaroo accident where the car had a rock hard pedal and no retardation,,

 

I have had this happen on my evo several times, usually due to severe ripples in the tarmac in heavy braking zones (old track with old surface) and my friend went out into the gravel in a porsche in the same corner..

I have also had this happen due to uneven/bumpy snow surface in the winter..

[Fat Boy]

Abs teaches you bad conventional braking skills. There again TC teaches bad launch technique and ESC needs different limit handling approach.

 

 

I do not understand why people do not fit balanced braking systems any more. It means that the abs intervenes sooner,and if it fails you are in trouble.

 

ABS braking skills: I'm not sure I'd say they're bad per say, just different. Likely the driver did not know that the tires were locked, at least initially. It's a car with fenders, you can't see the tires and the smoke is initially behind you. So all the driver would really know is that the car was not slowing at the desired rate. The driver would not expect lock-up to even be a consideration, so what would he do? Push harder on the pedal. If the anti-lock wasn't engaged (which any driver could feel...clearly it wasn't), then push harder and you'll slow down quicker. Well, not this time.

 

As far as the balanced braking system goes, it's normally biased towards the rear on an anti-lock car. On initial application, prior to long. load transfer taking over, a car will accept a significant amount _more_ rear bias than later in the brake zone (post long. load transfer). You can reduce braking distances by having that initial brake hit being rear-biased and then managing the rear lock electronically.

 

I'm a little confused about this particular failure. It seems strange that the rear anti-lock was somehow still engaged. I'm wondering what they were using as a reference speed if the fronts were not working. Maybe GPS? That seems like it would be a little too slow/erratic for such a task, but maybe not. If I were designing a road car system I don't think I'd separate the front and rears like that. Either it's all on or all off.

[Fat Boy]

I have had this happen on my evo several times, usually due to severe ripples in the tarmac in heavy braking zones (old track with old surface) and my friend went out into the gravel in a porsche in the same corner..

I have also had this happen due to uneven/bumpy snow surface in the winter..

 

Really common with road car systems that are overly intrusive. It's annoying as hell.

[Fat Boy]

I think the most likely (and safest) scenario is:

 - Front wheel sensor fails

 - ABS to front wheels disabled

 - ABS function retained on rear wheels

 

Simply reverting to non-ABS braking does not guarantee against a rear wheel lockup - especially with the pedal "mashed" to the floor. Eliminating rear brakes completely would dramatically increase stopping distance and in my mind constitute negligence.

 

The question is, "How do you retain rear wheel ABS with no meaningful reference speed (which is a combination of all 4)? Hell, man, just choose proper hydraulic ratios and kill the ABS on all 4. If that means 4 locked tires, then at least you're slowing. If you were trying to retain rear ABS without the proper sensors, I think it would be easy to end up with no rear brakes at all.

[gruntguru]

ABS does not require a reference speed (to compare to its own speed sensor) ie it can operate on a single wheel if necessary but certainly two wheels is plenty eg motorcycle.  It functions primarily by limiting wheel deceleration not speed - a sudden increase in deceleration beyond a certain threshold indicates lockup.

Edited by gruntguru, 26 July 2016 - 06:49.

[Greg Locock]

It detects wheel deceleration using the wheel speed sensor.

[gruntguru]

You're pulling my leg right?

[Greg Locock]

No. How do YOU think wheel deceleration is detected, if not via the tone wheels?

[Fat Boy]

I think he's alluding to using Longitudinal G or simply a max accel. for a given corner. For instance, if the LR wheelspeed is working, then we can get make a max positive and negative rotational  acceleration on that wheel to use as a surrogate for traction control or ABS. In reality, this is a really inaccurate way to do either and it's just not used.

 

On a motorcycle, T/C is done usually by using both wheelspeeds and possibly a GPS speed as well. In low gears when the front tire is in the air, the T/C is more of a gear dependent power mapping than actual T/C. The bike can tell if the front tire is off the ground with suspension travel/load or a pitch angle sensor. It will also be able to sense lean angle because speed on a motorcycle is very angle dependent. When you get that thing laid over, the tire radius changes dramatically.

[Greg Locock]

On a production car there's a tiny vehicle dynamics model that looks at the various sensor readings, decides what the car is doing, and brakes each wheel accordingly. For an abs the tone wheels and brake pedal pressure sensor and Ax are the main inputs.

[Fat Boy]

tone wheels?

[GreenMachine]

Ax?

[imaginesix]

tone wheels?

Reluctor wheel, probably

Ax?

Acceleration along the x axis (forward and back), probably.

[gruntguru]

 

It detects wheel deceleration using the wheel speed sensor.          

My workplace in the 1980's had the Bosch technical poster for ABS (among others) framed and displayed on the wall. 

[Greg Locock]

Tone wheel=toothed wheel
Ax=longitudinal acceleration.

Gg-too cryptic for me. Both wheel deceleration and wheel speed are derived from the distance measured by the tone wheel. Both are used in the abs computer.

[gruntguru]

Sorry. Must tone down the irony/sarcasm.

 

(have edited post #32 to eliminate potential source of confusion)

Edited by gruntguru, 26 July 2016 - 06:51.

[Catalina Park]

I know on the early ABS Commodores we would flat spot the front tyres by applying the handbrake first...

[GreenMachine]

I know on the early ABS Commodores we would flat spot the front tyres by applying the handbrake first...

Handbrake turns ABS off, at least on my Mazda. Just click it so the handbrake light comes on, and voila ...

[Catalina Park]

It does now, but it didn't then.

[mtknot]

Even after a crash like that, it amazes me how they're just able to bolt on new parts onto the tub like as if its an LMP1 or F1 car. 

[Lee Nicolle]

I would know immediatly if the fronts were locked, you feel it through the steering wheel. And I would back off the pedal, modulate, which it seems this person did not. Driver error, possible caused by panic and or not knowing the car without all the nanny items.

 

Though possibly these cars are set up different so as to use the nanny items.

They do create confusion. A woman I know who drives hard everywhere crashed a loan car,, forgot it had no ABS. Split second to wake up and hit another car with the rears locked up. Her own car was in for a recall.

Terrible or good driver depending on your view but the type who can get caught out easily.

[Greg Locock]

Switching from an ABS car to a non ABS one is extremely difficult. Different skillset.

[minime]

I would know immediatly if the fronts were locked, you feel it through the steering wheel. And I would back off the pedal, modulate, which it seems this person did not. Driver error, possible caused by panic and or not knowing the car without all the nanny items.

 

Though possibly these cars are set up different so as to use the nanny items.

They do create confusion. A woman I know who drives hard everywhere crashed a loan car,, forgot it had no ABS. Split second to wake up and hit another car with the rears locked up. Her own car was in for a recall.

Terrible or good driver depending on your view but the type who can get caught out easily.

 

Lee, do you ever run out stories for any situation?