RE: Prior Convictions: Heavy metal

[MikeDB1]

RacerMike said:

As I said in my previous post, you're confusing stability control and ABS. ABS is (as per the acronym) Antilock Braking System

As for your comments on track driving.....I'm afraid you need to do a little more research. ABS is very much used for track driving. In fact 99% of all GT4 and GT3 cars have motorsport ABS fitted. Having raced a GT4 car, I can tell you that it is very much a driving aid and both decreases braking distances on track and greatly increases consistency and tyre life through an hour long stint. Stability Control on the other hand is not permitted by any current motorsport regulations.

Firstly I'm not confusing anything. What I am pointing out is that thousands of rally cars do have an ABS (as defined by you) off switch fitted, or just drive around with the fuse flipped. Legal or otherwise, it's impossible to drive these cars properly with ABS systems enabled and everybody turns a blind eye to this.

Having driven at International rally level for most of my life, I can assure you that leaving ABS on would be both dangerous and slower overall through most corners compared with what I can achieve with my left foot. The Porsche systems you mention are derived from those developed in F1 some years ago and now banned in F1 and WRC. They have a wide selection of mappings which you can even vary corner by corner so are a long way from road systems. Effectively you are automating the driver function, but only once the driver has chosen the correct mapping for each corner. This is fine for the track, but not for a rally. Personally I'd like to see such systems banned in all motorsport but if fitted to a road car it would be acceptable as I could find a mapping I liked.

But looking at last Sunday, do you think any ABS system could have outperformed Ricciardo on his overtake ? My guess is any ABS system would have resulted in either no overtake or a crash. Computers are good, but only as good as those programming them and the data fed them from the sensors, none of which are going to be as good as world class drivers.

However this thread has gone a long way from its original roots. As you pointed out, short runs can be legally sold without the electronic aids and I still think Ford should sell a stripped out low weight 4WD Fiesta RS, possibly as a new brand or whatever is required to meet the letter of the law. The Mk2 RS1800, Capri 3100 and RS500 show this can be done profitably. Of course demand would far exceed the runs so the designs would need to change regularly to meet the rules, but I am sure this could be done - it just needs the same attitude Ford had when the first Mk3 Escort released was the XR3, not the slow shopping versions.

[RacerMike]

Just to be clear, there's no need to get heated here. I'm not intending on starting an argument. Far from it in fact. I'm (increasingly frustratingly) trying to explain a topic you have perhaps misunderstood and it's my intention to clarify and explain so you understand the comments others (including myself) have made regarding your suggestion that the spec you quoted would sell by the bucket load.

MikeDB1 said:

Firstly I'm not confusing anything. What I am pointing out is that thousands of rally cars do have an ABS (as defined by you) off switch fitted, or just drive around with the fuse flipped. Legal or otherwise, it's impossible to drive these cars properly with ABS systems enabled and everybody turns a blind eye to this.

I definitely didn't say rally cars had ABS. I actually said quite the opposite. Rally cars are SVA'd to enable them to be road registered. This means that they can be made legal to the SVA requirements which do not require ABS as a legality unlike series production cars which do. By all means find me a rally car that does has ABS, but I know for a fact that none of the WRC cars do. To be clear, there is also a difference between the laws that a manufacturer must abide by (which are vast and wide ranging) and the laws that you as an individual must abide by regarding your car's roadworthiness. You can (if you so wish) fit a switch to disable your entire ABS modulator, and I don't believe this would strictly be illegal. There isn't anything in the MOT Test that specifically asks the tester to 'look for an ABS defeat switch' however it must not show ABS faults if ABS is fitted.

MikeDB1 said:

Having driven at International rally level for most of my life, I can assure you that leaving ABS on would be both dangerous and slower overall through most corners compared with what I can achieve with my left foot. The Porsche systems you mention are derived from those developed in F1 some years ago and now banned in F1 and WRC. They have a wide selection of mappings which you can even vary corner by corner so are a long way from road systems. Effectively you are automating the driver function, but only once the driver has chosen the correct mapping for each corner. This is fine for the track, but not for a rally. Personally I'd like to see such systems banned in all motorsport but if fitted to a road car it would be acceptable as I could find a mapping I liked.

Not quite sure where I mentioned anything about a Porsche (?) system? The most widely used motorsport ABS system is manufactured and coded by Bosch. It utilises a standard production road car modulator as used widely across the automotive industry with only an ABS calibration in it. For the point of interest, this is actually a cut back version of the ABS logic that sits in a production car. A lot of the functional safety is deleted, a long with more road car oriented logics like split mu detection (the logic that detects when a driver is braking on a road that say has ice on one side and dry tarmac on the other) and usually EBD (electronic brake force distribution) since this is normally handled by a physical valve to change the bias manually.

The software parameters for each specific race car application are broadly similar, but key points like wheelbase, wheel circumference and vehicle mass are entered by a calibration engineer. There are also some additional tuning parameters that can be changed at the first application (but not subsequently by the team or driver). The driver than has a 12 position dial in the car which in layman's terms changes how 'much' ABS they get. What this actually does is change the point at which the system will start to intervene on wheel lock. Despite being fairly basic compared to a full road car application, it's actually very effective.

You could change it corner by corner if you really wanted to, but it wouldn't really do much as the only variable that changes for ABS during a race distance is tyre degradation. The adjustable positions effectively allow you to account for reduced grip due to older tyres (a less aggressive ABS target at this point helps to maintain stability on the brakes).

MikeDB1 said:

But looking at last Sunday, do you think any ABS system could have outperformed Ricciardo on his overtake ? My guess is any ABS system would have resulted in either no overtake or a crash. Computers are good, but only as good as those programming them and the data fed them from the sensors, none of which are going to be as good as world class drivers.

Actually, if F1 were allowed free rain to do this then yes, it would actually outperform Ricciardo. They more than likely actually do, do some form of brake modulation already. The rear circuit on an F1 car is brake by wire to allow them to blend between hydraulic brakes and regen. This enables them to keep the balance the driver experiences the same during a stop (as opposed to a balance that would shift forward towards the end of the stop due to the KERS's inability to regen as strongly at lower speeds). I wouldn't be half surprised if, much like the ERS during throttle applications that 'definitely isn't Traction Control Mr Todt' they choose to not regenerate the battery so much when the tyres are already 'achieving maximum regenerative capacity' as I suspect they would word it. In other words, I strongly suspect that as they detect wheel slip from braking on the rear axle, they start to bleed off KERS torque.

If F1 were to fully embrace ABS with modern technology it would be genuinely incredible. Free from the cost implications, they would be able to run a (rather expensive) real time tyre slip sensor (think of an optical mouse only bigger and more accurate) and some fairly trick hydraulics. Don't get me wrong, it would be tough to calibrate and pretty sketchy if it went wrong, but it would be possible to hold each individual wheel at absolutely the ideal tyre slip %age at all times from the moment the drive mashed his left foot into the bulkhead to the moment he released it. It would do this independently on each wheel regardless fo what the drive does, so it would most definitely be better than any human could possible be, unless they had 4 legs and 4 pedals.

MikeDB1 said:

However this thread has gone a long way from its original roots. As you pointed out, short runs can be legally sold without the electronic aids and I still think Ford should sell a stripped out low weight 4WD Fiesta RS, possibly as a new brand or whatever is required to meet the letter of the law. The Mk2 RS1800, Capri 3100 and RS500 show this can be done profitably. Of course demand would far exceed the runs so the designs would need to change regularly to meet the rules, but I am sure this could be done - it just needs the same attitude Ford had when the first Mk3 Escort released was the XR3, not the slow shopping versions.

As a few of us mentioned (which is why this thread has gone off topic (although I hope some people find it at least slightly interesting) unfortunately, modern economics don't work like that. If they made a 4WD 500run Fiesta RS it would almost certainly be north of £100k. The cost in developing a one off car for a major manufacturer is absolutely massive. Added to that the moral and legal responsibility for a major OEM like Ford to sell a car without safety systems it fits to all it's other cars (and has used a loophole to circumvent) and it's (perhaps unfortunately) never going to happen. There's nothing to stop you going out and building one for yourself of course.....

Edit to add:

I now understand why you thought I meant Porsche. By GT3 and GT4 I mean the FIA GT3 and GT4 regulations. So GT class cars that race in Blancpain, GTOpen and British GT. Porsche road car ABS/Stability control is Bosch Gen9, as per a good number of other OEMs like Land Rover, Jaguar, Mercedes, McLaren, Ferrari etc. The component and software is owned/developed by Bosch and then subsequently calibrated by Bosch engineers with the direction of engineers who work for the OEM (like me). There are some individual logics that are bespoke to each manufacturer, but the core programming and logic is the same across all vehicles with the same modulator (so across manufactures as well)

Edited by RacerMike on Monday 16th April 21:30

[MikeDB1]

RacerMike said:

I now understand why you thought I meant Porsche. By GT3 and GT4 I mean the FIA GT3 and GT4 regulations.

Yes I realised what you meant. I believe what you are referring to is the Bosch ClubSport system which I'm pretty sure was initially developed for, or at least on a, Porsche. This system does allow customer specific maps to be generated and stored and I'm quite sure all the top teams do so. However whilst based on standard hardware, it does have a much faster pulse time than on road cars. I'm not sure why road cars don't have this, possibly due to equipment lifetime concerns ? It sounds like you may know the answer to this so I would be interested to hear it. I'd certainly be less anti-ABS if the pulse time on road cars was reduced as it would give the system more finesse.

RacerMike said:

Actually, if F1 were allowed free rain to do this then yes, it would actually outperform Ricciardo.

I'm certainly not convinced on this. I take your point that the human body has a lack of actuation outputs, but it has far better sensors than any computer system, and a vastly superior non-AI based intelligence system. Research into such systems (not for cars) is my day job so I am aware of what such systems can and can't do. In maybe ten time then it will be different, but not yet.

RacerMike said:

As a few of us mentioned (which is why this thread has gone off topic (although I hope some people find it at least slightly interesting) unfortunately, modern economics don't work like that. If they made a 4WD 500run Fiesta RS it would almost certainly be north of £100k. The cost in developing a one off car for a major manufacturer is absolutely massive. Added to that the moral and legal responsibility for a major OEM like Ford to sell a car without safety systems it fits to all it's other cars (and has used a loophole to circumvent) and it's (perhaps unfortunately) never going to happen.

A brand new M-Sport Fiesta R2 2WD rally car is around £55k-ish and that's hand-built with competition engine and suspension, safety cage and the other stuff needed for rallying. I'm quite sure Ford could trigger a limited production run of a 4WD Fiesta without all the competition extras at well below this price, as they have done in the past with a long line of competion cars adapted for the road.

But as you say, the will to do this unfortunately doesn't exist. The problem of course is that Subaru have also lost the will to do this by ending their long run of WRX specials, which means the market is now unserved. A small manufacturer like Caterham simply isn't going to have access to the array of standard parts such a project needs at low cost, and major companies are never going to sell bodyshells in quantity so the end product is always going to be second rate.

RacerMike said:

There's nothing to stop you going out and building one for yourself of course.....

Apart from getting it insured. I have a well known self-built historic Italian rally car in the garage and it doesn't even need tax or an MOT now, but nobody will touch it for standard road insurance even though a bog standard road-going Focus RS easily beats it for acceleration, though it can of course still out-corner and out-brake the modern vehicle. Hence it has to be trailered to events whenever it appears.

Edited by MikeDB1 on Monday 16th April 23:15

Edited by MikeDB1 on Monday 16th April 23:18

[RacerMike]

MikeDB1 said:

Yes I realised what you meant. I believe what you are referring to is the Bosch ClubSport system which I'm pretty sure was initially developed for, or at least on a, Porsche. This system does allow customer specific maps to be generated and stored and I'm quite sure all the top teams do so. However whilst based on standard hardware, it does have a much faster pulse time than on road cars. I'm not sure why road cars don't have this, possibly due to equipment lifetime concerns ? It sounds like you may know the answer to this so I would be interested to hear it. I'd certainly be less anti-ABS if the pulse time on road cars was reduced as it would give the system more finesse.

The Bosch M4 (Clubsport) Motorsport ABS is a series production Bosch Gen8 modulator with a Motorsport ABS map. The fundamental logic is based on the road car ABS (not Porsche. All road cars have the same basic calculations and logics), but with modifications for motorsport. Compared to the calibration in a road car, there's very little in the way of car specific tuning. It takes an OEM and Bosch around 6 months to create a saleable ABS tune. The motorsport ABS takes 1 guy and a test day to set up as there aren't many parameters to change. Also, the teams aren't able to change anything. The software is free to download here ( Race ABS M5 Software Tool. There isn't any way to change the ABS maps, and having plugged into an Aston Martin Vantage GT4 with it the other week, I can tell you that even the very basic car data is locked out. To change the calibration you need a Bosch development ECU which is around £10k on it's own and needs specific development software tools and licenses unavailable to anyone outside Bosch or the OEM car industry.

The hardware in the M4 is a 4 channel, 2 piston pump modulator. Bosch M4 ABS This is actually pretty old hat now. Bosch Gen 9 Premium as used in higher end road cars has a 6 piston pump which means it's capable of more finely adjusting pressure. By 'pulse time' I assume you mean the pedal cycling you feel under foot? This has more to do with noise and refinement than actual performance. I think you possibly aren't aware of how a modern ABS software works, and perhaps perceive it as a 'dumb' controller that just cycles the brakes on and off? The cycling you feel is just the modulator varying the brake pressure to achieve a slip target, not 'speed' at which it's cycling as such. Any modern ABS works (in it's most basic form) like this.

• Driver hits the brakes and generates hydraulic brake pressure
• The ABS, through wheel speed measurement and accelerometers compares wheel speed to reference speed and calculates whether the deceleration of the wheel is plausible/calculates the amount of tyre slip
• As slip is calculated (comparing wheel speed to reference speed) pressure from the calliper applied by the driver is bled back into an accumulator in the modulator block to relieve brake pressure to bring the wheel slip back up to an optimum %age
• As the wheel recovers, the pump generates new brake pressure to feed back into the calliper to try and hold the desired slip
• This repeats until the vehicle is stationary

A modern (6 Piston) ABS controller can effectively hold the wheel in pretty much ideal slip for the entirety of the stop. So much so that if you look, you'll now see a perfect set of 11s rather than the old fashioned dotted lines of old mechanical ABS. The processing cycling time is around 1-5ms depending on the particular signal. 6 Piston pumps aren't really needed for Motorsport though as the use case for motorsport is incredibly straight forward and no where near as complicated as a road car. A race car is used in two conditions. Dry asphalt with slicks or wet asphalt with wets. It's fairly easy to get away with a basic tune, and the assumption is the driver will have a reasonable degree of ability. There are also no legal requirements! For that reason, you can't put a Bosch Motorsport ABS unit on a road car as it won't meet the required performance and stability targets stipulated by the law.

MikeDB1 said:

I'm certainly not convinced on this. I take your point that the human body has a lack of actuation outputs, but it has far better sensors than any computer system, and a vastly superior non-AI based intelligence system. Research into such systems (not for cars) is my day job so I am aware of what such systems can and can't do. In maybe ten time then it will be different, but not yet.

ABS doesn't require anything particularly clever. It's a simple PID control system. Science tells you what tyre contact patch force you'll achieve at what tyre slip percentage and PID controllers are incredibly good at targeting a number to aim at. The only error states in a road car ABS system are in the sensing of wheel slip. It has to be estimated from reference speed (which can be unreliable) and even then, you never really know what the car is doing for sure. If you have a tyre slip sensor, you know for sure what the tyre contact patch is doing far better than a human does. Even without this, a modern ABS controller is far better on a dry surface than a human. I don't care how good you think you are, let's go out and do 50 stops from 150mph and we'll see how many times you win vs the ABS.

MikeDB1 said:

A brand new M-Sport Fiesta R2 2WD rally car is around £55k-ish and that's hand-built with competition engine and suspension, safety cage and the other stuff needed for rallying. I'm quite sure Ford could trigger a limited production run of a 4WD Fiesta without all the competition extras at well below this price, as they have done in the past with a long line of competion cars adapted for the road.

You've kind of answered your own question there. That's £55k for a 2WD Fiesta. What are the competition extras on an R2 car exactly? A cage, no interior, some harnesses and a pair of competition seats which probably, at most costs £5k. So you're paying £50k for a remapped 1.0l Ecoboost Fiesta. What you've suggested (4WD NA with no ABS) is basically an S2000 car. Which a quick search online suggests is around £100k second hand ( https://rallycarsforsale.net/ads/ford-fiesta-s2000...

MikeDB1 said:

Apart from getting it insured. I have a well known self-built historic Italian rally car in the garage and it doesn't even need tax or an MOT now, but nobody will touch it for standard road insurance even though a bog standard road-going Focus RS easily beats it for acceleration, though it can of course still out-corner and out-brake the modern vehicle. Hence it has to be trailered to events whenever it appears.

Eh? Getting a car taxed and insured in the UK is ridiculously easy. There are literally hundreds of people out there running all sorts of crazy stuff on the road. Including a guy on here who's put the floorpan of a Mitsubishi EVO into a Citroen C2. That's been road registered, and drives round all the time from what I can see....

Edited by RacerMike on Tuesday 17th April 01:21

[MikeDB1]

RacerMike said:

MikeDB1 said:

A brand new M-Sport Fiesta R2 2WD rally car is around £55k-ish and that's hand-built with competition engine and suspension, safety cage and the other stuff needed for rallying. I'm quite sure Ford could trigger a limited production run of a 4WD Fiesta without all the competition extras at well below this price, as they have done in the past with a long line of competion cars adapted for the road.

You've kind of answered your own question there. That's £55k for a 2WD Fiesta. What are the competition extras on an R2 car exactly? A cage, no interior, some harnesses and a pair of competition seats which probably, at most costs £5k. So you're paying £50k for a remapped 1.0l Ecoboost Fiesta.

Actually it's £30k of extra bits, plus of course you are throwing away almost a whole Fiesta apart from the body and a few other bits which wouldn't happen if Ford built the car from scratch.

http://www.m-sport.co.uk/images/New_R2_Sales_Docum...

But I don't think people want an R2 car on the road (300 miles between services :-) as that's well OTT, so I still maintain Ford could produce a performance 4WD Fiesta at a decent price, as indeed was rumoured in the press some years ago.

RacerMike said:

Eh? Getting a car taxed and insured in the UK is ridiculously easy. There are literally hundreds of people out there running all sorts of crazy stuff on the road. Including a guy on here who's put the floorpan of a Mitsubishi EVO into a Citroen C2. That's been road registered, and drives round all the time from what I can see....

Well I would be interested to know who's insuring him, but I was quoted over £2500 to insure this car for what would be a few hundred miles each year so a trailer works out cheaper. My modified Impreza is £226 per annum so it isn't due to any lack of driving experience (45 years) or bad driving record (45 years no-claims). But if you have some suggestions, I'll have another go and let you know the quotes I get.

[MikeDB1]

RacerMike said:

By 'pulse time' I assume you mean the pedal cycling you feel under foot? This has more to do with noise and refinement than actual performance. I think you possibly aren't aware of how a modern ABS software works, and perhaps perceive it as a 'dumb' controller that just cycles the brakes on and off? The cycling you feel is just the modulator varying the brake pressure to achieve a slip target, not 'speed' at which it's cycling as such.

I don't mean the pulsing that is fed back to the pedal.

I'm also well aware of how vehicle electronics works, having worked as an expert assessor on several EC funded projects under FP4 and FP5 in this area, and will in future be reviewing some projects for automated driving (hence the interest in AI).

Pulse time is determined by the modulation frequency of the controller driving the pump, basically the fastest time the controller can act to any change of input. You may call it something different now but it was definitely called that back in the early 2000s.

[anonymous-user]

MikeDB1 said:

The Mk2 RS1800, Capri 3100 and RS500 show this can be done profitably.

Eh? Those cars went out of production in the 1980's which, in case you haven't been keeping up was about 30 years ago!

I developed the Mk1 Focus RS for Ford in 2002, and that cars modification were relatively limited (turbo engine, different suspension etc). Ford lost over £5k on every one of the 4500 they sold......

To put a 4wd powertrain into a 2wd body-in-white is simply not going to happen except for the occasional "built in a shed" one-off, sorry.

Regarding ABS, just because "Rally cars" have an off switch means nothing. I developed GrpN, GrpA and WRC cars when working at Prodrive and allthough all there cars had an "MOT" they were far from Road Legal if you were to apply the letter of the law. As you mentioned, a blind eye was turned to that at the time (because it was good UK business, as motorsport still is, and because the 'production' volumes were low, and because the modifications were technically done post registration (by Prodrive, rather than IHI)

The biggest difference between 'Race' and 'Road' ABS is that road ABS has "maintain yaw stability" as it's highest rule. In any situation where brake force is radically different side to side or front to back, the system "picks low" and reduces the high grip wheel brake force to match the low grip one. This is because normal road drivers need help to maintain stability (and hence directional control) during panic and accident maneuvers (and hitting objects straight on, rather than sideways saves lives (because side impact accidents are very difficult to mitigate against). So, in a road car, when the ABS kicks in to prevent a wheel from decelerating, it actually extends braking distances. Motorsport systems have a lower priority on stability, and usually the "Dry/Wet" adjustment dial is changing both the slip target and the yaw stability threshold. In effect, it relies on the driver using the Handwheel to maintain yaw stability whilst the system tries to get the best overall stop by keeping all tyres at their individual optimum slip points, irrespective (well, nearly) of the yaw torques created whilst doing so!


IME,Rally cars woudl stop "better" with a modern ABS system, but the biggest hurdle is determinnig the optimum slip ratio for any given set of road conditions, and to some degree determining the true vector of the vehicle. Again at Prodrive we messed around with hugely expensive on board doppler radar systems to try to monitor the true vehicle vector (in that case, we were trying to optimise traction with the Active Torque Dynamics system, but the end effect (max tyre adhesion) is the same really.....

[anonymous-user]

MikeDB1 said:

RacerMike said:

By 'pulse time' I assume you mean the pedal cycling you feel under foot? This has more to do with noise and refinement than actual performance. I think you possibly aren't aware of how a modern ABS software works, and perhaps perceive it as a 'dumb' controller that just cycles the brakes on and off? The cycling you feel is just the modulator varying the brake pressure to achieve a slip target, not 'speed' at which it's cycling as such.

I don't mean the pulsing that is fed back to the pedal.

I'm also well aware of how vehicle electronics works, having worked as an expert assessor on several EC funded projects under FP4 and FP5 in this area, and will in future be reviewing some projects for automated driving (hence the interest in AI).

Pulse time is determined by the modulation frequency of the controller driving the pump, basically the fastest time the controller can act to any change of input. You may call it something different now but it was definitely called that back in the early 2000s.

Limit factor in all cases for Pressure cycle time is the wheel/tyre/driveshaft (and even powertrain) rotational inertia and the current co-efficient of Friction. This is because the wheel/tyre must be re-accelerated back up to speed (low slip) by the road itself. By comparison the task schedule and code execution rasters in any modern processor are many decades faster!

(interestingly, EV's are opening new opportunities, with the high frequency, bi-directional torque modulation (typically over 50KNm/sec at the eMachine) meaning we can now control wheel slit linearly, or use that eMachine to re-accelerate the wheel/tyre as necessary, meaning faster (two mode) control cycles ;-) )

[MikeDB1]

Max_Torque said:

MikeDB1 said:

The Mk2 RS1800, Capri 3100 and RS500 show this can be done profitably.

Eh? Those cars went out of production in the 1980's which, in case you haven't been keeping up was about 30 years ago!

Yes I do realise that. Subaru took over the performance road car mantle from Ford in the 1990s but now they have abandoned it, it needs some car manufacturer to take up the challenge once more. Let's hope the Koreans have a go once they eventually win the WRC.

Max_Torque said:

I developed the Mk1 Focus RS for Ford in 2002, and that cars modification were relatively limited (turbo engine, different suspension etc). Ford lost over £5k on every one of the 4500 they sold......

Yes that was well documented, but surely that shows that Subaru were doing a better job with the Impreza which limited Ford's list price.

Max_Torque said:

IME,Rally cars woudl stop "better" with a modern ABS system, but the biggest hurdle is determinnig the optimum slip ratio for any given set of road conditions, and to some degree determining the true vector of the vehicle. Again at Prodrive we messed around with hugely expensive on board doppler radar systems to try to monitor the true vehicle vector (in that case, we were trying to optimise traction with the Active Torque Dynamics system, but the end effect (max tyre adhesion) is the same really.....

Sounds a very interesting system, but it emphasises that maximising corner speed isn't just a function of the brakes. Maybe an ABS rally car would stop better but would it corner faster overall compared with how a top driver would slow it down ?

And might the expense of it have been better spent on a better driver ? You don't give a date as to when this was but as we can see in WRC nowadays, there is actually a huge variation in driver ability (i.e. first name equals Sebastian or not :-), despite comments in F1 that drivers can be plugged in like light bulbs.

[MikeDB1]

Max_Torque said:

(interestingly, EV's are opening new opportunities, with the high frequency, bi-directional torque modulation (typically over 50KNm/sec at the eMachine) meaning we can now control wheel slit linearly, or use that eMachine to re-accelerate the wheel/tyre as necessary, meaning faster (two mode) control cycles ;-) )

I totally agree on this one. I did do a paper exercise on whether a custom home built EV could win single venue rallies as there's usually time available to change the battery packs between stages and it's actually getting very close to being the case. There's even a place in the US selling all the major sub-systems for both 2WD and 4WD EV custom builds.

But still a long way off being able to carry around a whole day's charge no matter how much energy recovery you do.

[RacerMike]

Max_Torque said:

Limit factor in all cases for Pressure cycle time is the wheel/tyre/driveshaft (and even powertrain) rotational inertia and the current co-efficient of Friction. This is because the wheel/tyre must be re-accelerated back up to speed (low slip) by the road itself. By comparison the task schedule and code execution rasters in any modern processor are many decades faster!

(interestingly, EV's are opening new opportunities, with the high frequency, bi-directional torque modulation (typically over 50KNm/sec at the eMachine) meaning we can now control wheel slit linearly, or use that eMachine to re-accelerate the wheel/tyre as necessary, meaning faster (two mode) control cycles ;-) )

Not entirely true. Drag Torque Control has been in ABS for at least a decade now and during an ABS stop the engine very effectively used to spin the wheels back up to road speed. It's quite tough to calibrate (especially on a 4WD car), but it's very effective.

I know we've also had this discussion before elsewhere, but I'm not sure I agree with your belief that a driver will stop better without ABS. It may be true in some very specific cases (if for instance you had to brake at a brake marker and the driver had a chance to practice), but in absolutely any real world situation, ABS will do a better job more of the time. With modern systems able to work almost exactly like a professional driver would during threshold braking as well, I'm not convinced even driving skill would beat it.

Working in ABS and Stability Control it never fails to amaze me how many good engineers fail to update or understand the developments in modern ABS. Many other dynamics engineers still seem to see it as this thing that just stops wheels locking. I'm not suggesting you're like this, but I think it's worth realising that the ability of the systems is every increasing. The next generation systems are another level entirely.

[anonymous-user]

RacerMike said:

Not entirely true. Drag Torque Control has been in ABS for at least a decade now and during an ABS stop the engine very effectively used to spin the wheels back up to road speed. It's quite tough to calibrate (especially on a 4WD car), but it's very effective.

The problem is the response time of the powertrain, even at say 3000 rpm, there are just 25 firing events per second, add in things like manifold filling and lag of forced induction systems, and as you mention it is very difficult to get precise control of wheel torque in real time. It does work in really slippy conditions (ice. mud, gravel etc), where pretty much any additional wheel accel is a significant benefit but an EV has none of those latency issues, and typically controls eMachine fundamental torque at well over 1 KHz and can furnish wheel torque slew rates of easily over 200KNm/second !

RacerMike said:

I know we've also had this discussion before elsewhere, but I'm not sure I agree with your belief that a driver will stop better without ABS. It may be true in some very specific cases (if for instance you had to brake at a brake marker and the driver had a chance to practice), but in absolutely any real world situation, ABS will do a better job more of the time. With modern systems able to work almost exactly like a professional driver would during threshold braking as well, I'm not convinced even driving skill would beat it.

You might not be referring to me, but i am firmly in the "you can't really beat ABS" camp! The last time i was able to reliably, ie not just in perfect repeat conditions, out-stop ABS was in around 2007, before the widespread introduction of yaw sensing and stability control. Before that time the ABS had to always er on the side of stability, leaving a large margin for the unknown, because the system could not measure stability criteria in real time. Since the introduction of on-board direct,and real time yaw and acceleration sensing, the system now measures stability at all times, and hence can threshold break all wheels right up to that level of stability, making it pretty much unbeatable. Add in adaptive road friction learning and by the time the human has learnt, so has the computer!

RacerMike said:

Working in ABS and Stability Control it never fails to amaze me how many good engineers fail to update or understand the developments in modern ABS. Many other dynamics engineers still seem to see it as this thing that just stops wheels locking. I'm not suggesting you're like this, but I think it's worth realising that the ability of the systems is every increasing. The next generation systems are another level entirely.

I actually had a good demo of the powers of modern systems with a bit of code that could be run in our ATD car of the time that would make it effectively a one wheel drive, one wheel braking car! Lock a single front wheel, or apply full engine torque to a single rear wheel and the rate at which the car would swap ends was often eye watering (sometimes literally when you banged your head on the roll cage as it went round....)