84 replies to this topic

[desmo]

I would contend that any ban on traction control is unenforcable with the current engine and transmission managment systems in place in F1. Unless you mandate points and distributor ignitions, carbs, and throttle cables there will always be a way to detect wheelspin and modulate power delivery. Anyone disagree?

[Ray Bell]

They could probably do it via telemetry and hide it that way... You're probably quite right.

------------------
Life and love are mixed with pain...

[Matt Davis]

I thought transmiting data back from the pits to the cars to controll part of the car woz ileagal.

[desmo]

So is traction control. What's your point?

[Art]

Desmo.

If they are using the ignition to controle traction controle. Recording the sound at the start of the race of individual cars. And putting it on a Spectrum Analizer could detect variations in engine sound. Giving away ignition controled traction controle.

Art

[SalutGilles]

But couldn't the fly-by-wire system get the job done? it would only allow the maximum-non-wheelspin input in a specific gear, wouldn't it?

[Ray Bell]

There would have to be more to it than that.. speed sensors, maybe suspension load sensors... tricky gear. Wouldn't like it to glitch out in the middle of a 120mph sweeper!

------------------
Life and love are mixed with pain...

[desmo]

Art, I am fairly certain that the race engineers for the top teams would have little difficulty coming up with a power modulation system that would not produce a tell-tale audio signiture. And just imagine trying to write the regulations that would define EXACTLY what audio signitures would be legal against those that wouldn't!

Ray, I don't think it would have to be that complicated. Let's say that any increase in revs under power that exceeds the expected climb from a hooked-up drive is interpreted as wheelspin. No sensors needed beyond any that are already there to provide data to the otherwise legal engine management system.

[Art]

Desmo.

Other than variable power or using a slipper clutch are you going to gain controle of traction?

Art

[IndyIan]

I think taking away the sensors would be enough to eliminate traction control use except for the start. Having preprogrammed engine mapping for each gear would be fine on the start but the traction used for cornering during the race would make these systems not that useful.
The start would still be ruined by TC but not the actual race. This might be the best compromise available. It could also lead to different fuel strategies with drivers choosing to start with little fuel to gain places at the start.
I personally hope they can get rid of TC altogether and the less driver aids the better. I want to see drivers race instead of software.

As far as detecting TC by audio analysis I think that it would work provided the FIA gets all the telemetry from the cars. When the engine note changes and the drivers foot doesn't move that is proof of TC.

[desmo]

IndyIan,

How would you write the regulations in a completely non-subjective way to define what exactly costitutes proof of TC intent, as opposed to simply a change in the mapping of the ECM in response to one of the many variables that it must monitor to optimise engine performance? If you can write your idea in a completely foolproof and unambiguous manner that is not open to multiple plausible interpretations, I'd love to see it. I know of no definition of "engine note" that is precise enough to make it a useful term as far as drafting technical regulations.

[This message has been edited by desmo (edited 04-04-2000).]

[Art]

Desmo.

With all the stink about safety and cheating there is a simple answer. Go to 2 Liters and give them a free hand at design.

Art

[desmo]

Art,

I'd probably go all the way down to 1.5L just to keep Max out of the game a little longer.

[Ray Bell]

A 405 bhp 1.5-litre V10 turning 19,000 rpm would be nice!
Getting back to the argument, desmo seems to have a good handle on this. But one thing that shines through is that everyone seems to be thinking about legislating via noise.
That's not the point. The noise recording and analysis would point to the discrepancy, then it would be up to the computer experts to look at the software.
The legislation is there already, what is needed is a means of identifying the cheats.

------------------
Life and love are mixed with pain...

[PDA]

perhaps a posiible solution would be to reverse the current thinking which implies what one may no do, to positive ruling which states what one can do. Making sure that foot throttle to manifold throttle is linear and that manifold throttle opening may not be other than in a direct and constant relationship to foot throttle position would go a long way to solving the problem. This is supposed to happen now, according to the rules, but seems not to be. So mandate the code to be used to control throttle opening. Is this over simplification? or does it start along the right track?

[desmo]

These are the regulations as they currently stand:


5.7 Throttle control :

5.7.1 ) Other than the specific exceptions mentioned below in 5.7.2, there must be a fixed relationship between the position of
the throttle pedal and the engine throttles. This relationship need not be linear but the position of the engine throttles may not be
influenced by anything other than movement of the throttle pedal when operated by the driver.

This relationship must remain fixed whilst the car is in motion subject only to Article 8.3.

5.7.2) The relationship between the throttle pedal and engine throttles may alter during one or more of the following operations
:

- idle control ;

- stall prevention ;

- gear changing ;

- car speed limiting.

I think they are quite reasonable and in my reading make TC via the ECM modulating the throttle to aid the driver during starts and acceleration clearly illegal.

[Ray Bell]

Isn't traction control, like rev limiting, done via ignition interruptions? And, with fuel injection, throttle controls don't have to be altered to cut off the flow of fuel - throttle controls the airflow, the injectors the fuel flow...
But I agree that positive legislation is better: "you may" instead of "you may not"

------------------
Life and love are mixed with pain...

[PDA]

Good points Ray, but air flow/fuel flow are closely related, and as pointed out earlier, if fuel flow is drastically decreased while air flow remains high, the weak mixture would lead to overheating (or melt down). Your suggestion of ignition interruption or retardation seems a more likely method, and this is no doubt a lot more sophisticated than the old "switch off one or two cylinders" used in 92/3.

One thought, the data logging and telemetry is of no use during a race, so ban its recording and collection. That way, they would not be able to collect the speed/engine speed data needed to calculate whether wheelspin is taking place etc.

[desmo]

I think you could achieve TC using an open loop system that simply compares the rise in revs to a map of expected rise if the rear wheels are hooked up. If the rise in revs exceeds the preset threshold the ECM retards the timing and power is modulated. No sensors or telemetry needed.

[davo]

Desmo:

As you say it would not seem to be too hard to have a spark timing variation based on engine speed acceleration. It would need to be a little more advanced than the old mechanical advance in distributors.

A system which looked at the gear, speed and acceleration could be fairly easily tuned, the computers used could certainly be mapped to a quite complex approach. It may not be perfect but I would think it would not be hard to out perform a human.

About the only detraction from all of this is if it was open loop rather than closed loop control than getting an optimised map would need quite a bit of testing.

On an entirely separate subject which team have just spent time (and probably a lot of money) practicing starts up and down a drag strip? I bet their drivers really nail the next GP start.

[IndyIan]

Desmo:

"I think you could achieve TC using an open loop system that simply compares the rise in revs to a map of expected rise if the rear wheels are hooked up. If the rise in revs exceeds the preset threshold the ECM ******* the timing and power is modulated. No sensors or telemetry needed"

This would work on a start or any straight line acceleration but during the race the biggest advantage of TC is acceleration out of the corners while there are still side forces on the tires or in the middle of the corner. I think you would find a simple system like you've suggested would only make a difference coming out of very low speed corners on to a straight.

I think sonic detection of TC would work by recording samples of the car accelerating down a straight where wheelspin is not an issue and compare this sound to the sound generated coming out of corners where TC would be activated. These two sounds could be recorded every session so no written definition is neccessary. Also having the teams telemetry of the drivers inputs would make this task easier

I think you could probably ask the corner workers right now which cars use TC because I'm sure they sound different when its activated unless the TC system applies the rear brakes but that kind of TC componentry would be very hard to hide.

[Art]

As far as pre recorded audio signature it isn't necessary. An audio pick up could be mounted in the engine room of each car with a telemetry Xmitter. Receivcer's could be mounted in each turn and a central recieving tower Antena connected to a Spectrum Analizer for each car in a sound truck. Any radical variation in audio and a black flag and instant DQ.

Art

[desmo]

The problem with using audio as the sole criterium for regulatory desicions is that of defining exactly what is the threshhold between legality and illegality. The situation reminds me of the famous quote by a Supreme Court justice who said (and I'm paraphrasing here): "I can't tell you the difference between art and pornography, but I sure know it when I see it".

This is analgous to the situation raised by attempting to define TC using an audio sample. You can't tell me the EXACT difference betwen what is legal and what isn't but you'll know it when you hear it. I challenge you to write a set of regulations that define precisely where you draw the line based solely on audio.

I know it sounds simple, but I don't believe it to be possible using strictly objective criteria, which one obviously must when deciding who does or doesn't get DQed. Sometimes things that sound simple upon casual examination are not only difficult but impossible to do when put to a standard of proof.

Bernie once commisioned a Daimler-Benz engineer who was the most qualified in the world to develop an extremely accurate in-line fuel flow meter. Sounds simple enough, right? It turns out it was impossible to do to the degree of accuracy required to use it as a data acquisition device for deciding legality or illegality. Things get tricky real fast when it becomes necessary to abandon intuitive truth and work to a higher objective standard.



[This message has been edited by desmo (edited 04-05-2000).]

[Art]

Desmo.
The rule? 1. Any iratic deviation from the video taken of the Spectrum Analizer of a legal car accelerating from the start or out of a turn during the race will be considered illegal. And will br disqualified.

Art

[desmo]

Art,

You prove my point. Erratic is not an objective, but a subjective term. You cannot quantify whether something is erratic. At what quantifiable threshhold does something become erratic? What exactly is a spectrum analyser? Like the in-line fuel flow meter, it sounds easy but it is not. You think interpretation of the rules is open to accusations of political influence now, at least the rules are defined in objective terms. Introduce subjective terms and the whole thing will digress into mayhem.

[Art]

Desmo.

The Spectrum Analizer is a high grade Oscilloscope. A metal box with a cathode ray tube and numerous switches to set it up. the audio being analized is projected on the tube for visual observance. They have a reference storage unit and you can compare what is and what should be. These things are so good that if the microphone picked up the driver breaking a little air the picture would be in front of you. Other weise you can see sound.

Art

[Ray Bell]

Subtleties, like the facts, escape poor old
Art at times. Desmo has clearly explained the position, but Art keeps hammering with his ideas. It's okay if all you have to do is keep exactly 4'8.5" between the rails so the train won't jump off, but legalities are based on the exact meaning of words...

------------------
Life and love are mixed with pain...

[Sash]

I don't see why a lot of people are asying interoduce a throttle cable. You can buy aftermarket engine management systems for your road car that has traction control built in. It just ******* the timing to lose engine power. Cutting the ignition or varying fuel is definately not the engine friendly way of doing this.

IMO I think all that engine noise detection is crap. No way would it work practically. Maybe in theory only.

[davo]

From Atlas F1 News Sat 1st Re Benetton: "On Thursday the test programme moved to the Santa Pod drag strip where the drivers were scheduled to practice starting techniques for the entire day."

REALITY CHECK!

Think about what this news demonstrates. Benetton couldn't possibly fit some standing starts into a normal days on circuit testing? (Where the $s and time can be put to some other development uses).

Instead that went to the time and commitment of a full days testing exclusively for this????? Somehow it doesn't add up. Benetton have been a team that has historically strayed further down the "creative" end of the rules spectrum, than some.

It leads the perpetually suspicious to believe they might have been developing/testing/mapping a traction control system.

Lets see what the next few race starts bring.

[Art]

Yes poor old senile Art doesn't understand the facts! He just keeps hammering away with his stupid unfounded idea. If you gentelmen would stop at your local Library and get a book on Spectrum Analizer's. Read and understand it then comment on the subject. And 4'8.5" is all it takes to keep the train on the track. Don't try and play Railroading with me you have the wrong guy. Pakton Technologies Pty Ltd may have a Spectrom Analizer in there lab visit them and they may tell you the ability of this electronic wonder.

Art

[mono-posto]

The problem is unenforcable because the teams pay more and hire smarter engineers than the FIA does!! ;) They will always be one step ahead.

I think Mr.Mosley and his technical team have been hell bent on reducing lap times from a saftey point of view, which is all well and good. But they have been focusing on doing so through the constant limitaion of traction. And as we can see this is the wrong direction to go as lap times have held or even gotten quicker in some areas!

I think a much more effective and enforcable method is by limiting power via displacement and/or air flow regulation.

I personally feel that the cars should return to slicks, be given wider tracks and allowed to develop TC. But have displacement limited to reduce speed. Not only would this have the desired effect of slower lap times the cars would be leaving the track alot less!!

And that is what I don't understand about Mosley's safety campaign. How can a car be safer if it has such little grip that it spins off so easily?

[This message has been edited by mono-posto (edited 04-06-2000).]

[davo]

Art - certainly not an attempt to sully your faculties or your insights. I think you're right that the audio would be a fairly obvious give away. Certainly comments from those track side of "off" sounding exhausts when TC might be most useful are a give away.

However how you actually define the rule is the challenge. The 6cm ride height farce came from a poorly defined rule, in the era of "gentlemen lift your skirts". You could see the ride height on the track but you got to measure the ride height in the pits.

Teams ARE cheating. The technical ease of achieving TC, the "off" sounding exhausts, the testing teams do.

Rules which are unenforced, erratically enforced, or indeed unenforcable are BAD for the sport.

My utopia would be for rule making to take place where each team owner and just one of their engineers was locked in a conference room the day after the last race of the season.

Feed them Maccas for every meal
Give them toilet breaks
but don't release them until there is a set of rules for the next year

Majority vote not the useless unanimous agreement that we currently cannot see the detail of.

We would have rules that worked.

[PDA]

Mono-Posto - you are mistaken. TC was banned, along with ABS, to reduce driver assistance, and consequently increase depenance on the skill of the driver.

[Art]

MN.

This was not originally posted by Art.

Art

[mojo jojo]

The FIA should:

hire more CS majors. Before the season starts, they check and approve the code from the engine management systems. It would be the responsibility of the teams to convince the inspectors that they are abiding by the rules. This will take some time, so maybe one person for every team, or every two teams. The inspector approves the code and the machine level instructions that will be used on the car. At each race the checksum of the software loaded in the car would be checked. This is like the process that a zipping utility uses to check for errors after decompressing a file. If the code is changed in the slightest way, the checksum will be different.

With the many millions that F1 rakes in every year, surely they can afford a few hundred thousand to hire full time technical staff to police the teams. Say, 4 or 5 groups for the entire field. Each group has at least one CS, EE and ME person. The ME guy wouldn't be so important, but the EE and CS people would make sure each car's electrical systems and engine management software conform to the rules.

[Ray Bell]

Now this thread comes up with an answer that might be heading in the right direction... I did say 'might' - we still have a ways to go.
My reference to the distance between rails, by the way, was to explain to Art that this was a more complex question.

------------------
Life and love are mixed with pain...

[Art]

4'8.5" doesn't always keep a Train on the track!

Art

[Art]

No one has come up with Scientific proof that the Spectrum Analizer can't do the job. In the Scientific comunity I don't have to prove i'm right you have to prove I'm wrong.

Art

[desmo]

Art, You are obviously not familiar with the rigors of the scientific method. When you present a hypothesis such as yours, it can't simply be a purely speculative position such as: "there'e this super osciloscope thing called a spectrum analyzer and it can tell if someones using TC or not." It is incumbent upon the claimant to prove such a statement to be true and then it must pass the muster of peer review whereby others in the field try their counter-arguments to find a flaw. Read through the post again. I have put questions forward that I think must be addressed before such a totally unsupported claim such as yours can be taken as fact. I hope all this is not lost on you.

[This message has been edited by desmo (edited 04-10-2000).]

[Ray Bell]

Desmo - how's the head?

------------------
Life and love are mixed with pain...

[Darren]

Hi Art,

I'll give you one really simple scientific reason why the spectrum analyser approach wouldn't work the way you think it would: reference points and controls. To achieve the degree of accuracy in detecting traction control-modulated exhaust notes necessary to make reliable judgments, you'd need every car impounded by the FIA on the Thursday of race day; the FIA would have to guarantee no traction control was on the car; the FIA test driver would have to run laps in each car to get the right graph for each car(composed of an average of results from the laps run); the FIA would have to return the car with its traction control gear fitted; and all the real drivers will take odd lines into corners, lift off to avoid slow cars, push wide when they lose gears, left-foot brake and throttle jockey, and just generally drive differently to each other, causing all sorts of different engine notes. So you can't really establish a reliable reference point in what is essentially a wildly dynamic system. And logistically, it's not an option of first choice.

[Art]

Desmo.

I will make one last statement on the subject. The moving picture of acceleration on the scope looks like the constant upward grade of the first hill on a roller coaster. But all of a sudden this constant upward grade levels out. Then continues it's constant upward grade while accelerating out of each turn. Does this show you that monkey buissness is going on? Maybe I'm like the Texas redneck with a new Idea?

Art

[Art]

PS.

The first two cars in yesterday's race sure didn't have time to play throttle jockeys. And run all over the track to change there engine note.

Art

[desmo]

Ray, Heads fine today. If you'd asked me yeesterday... Pretenders were being unremittingly masochistic at the billiard table Friday night. Good thing I rode my bicycle!

Art, I never said you couldn't tell who might be cheating by analysing a audio sample (frankly I don't know), just that I am pretty certain you could never prove it. The difference is substantial.

[Art]

Desmo.

The reason I brought the subject up in the first place. A few years ago while trying to work some Islands in the south Pacific on my Amateur Radio Station. I was getting an intermittent noise that was wrecking every thing. So I called the Electric Co. they brought the analizer out with a small antenna hooked to it. They drove around for a couple of hours and returned and told me the noise was from an aquarium heater 3/4 of a mile away. No more noise and I believe in the Spectrum Analizer after being told what all it could do by an sound engineer.

Art

[Art]

Desmo.

It sounds like you were on the Booze over the weak end?? I use to be the (biggest drunk in town). But havn't had a drink in 22 years this March. And I would love to have all the money that I have pissed down urnals. Get away from it before it is to late.

Art

[Ray Bell]

You sure do seem to have had an interesting life - the quiet life on a South Pacific island, the noise of a North Atlantic railway (sorry, railroad), and undoubtedly a lot more in between.
You're a formidable opponent here, too, with your stoic refusal to give in on issues in which you have a deep belief. desmo has learned to his frustration that you are not a man to be toyed with. And no doubt the Nurse has learned the same over the years.
They obviously built you tough all those years ago, and the Nurse must be made of the same stuff to keep going all these years.
Does she have the key to your liquor cabinet, by the way?
I must say that I agree with you and reinforce you advice to poor young desmo. This is not the sort of treatment he's used to, and he would've given up some time ago had he not seen the need to show you the respect he has.
Undoubtedly he was pleased to see you would not be posting any more on this thread, showing that other quality that's so desirable, the ability to give up when you're licked.

------------------
Life and love are mixed with pain...

[This message has been edited by Ray Bell (edited 04-10-2000).]

[Art]

Licked My Ass.

Anything newer than leaf springs and wooden spoked wheels and you guys are lost.

Art

[Art]

Look up Amateur Radio Call Book on the net. Then type in NX3L. You will find out who I am and that I hold the highest class license an Amateur can hold.(EXTRA CLASS)

Art

[desmo]

Art, You're driving me to drink! Don't worry, the definition of impossible may mention doing fine scrimshaw while hung over. Best wishes now, I'm off to adjust the wooden spokes and leaf springs on my velocipede.