Time gained thanks to traction control?

[henrykleinespel]

Hi!
First post on this forum

I'm currently working on traction control for my "2nd year college project".
I found many bits and pieces of information about the system on the internet, but I didn't find anything about the time gains obtained with traction control.

So here's my question:
What was the typical time difference between a "TRC lap" and a "no TRC lap" ? (in F1 obviously).

This would really help me! (if I can't tell the people who will be grading me, what TRC brings in terms of performance, they'll not take me seriously )

[raymondu999]

Hey. Welcome to the forums

I don't think you can quantify that, because it really depends on driver feel. If a driver has bad feel and can only access 20% of the traction he has, then there will be huge gains. If you're talking of an exceptional driver, then he can get quite close to 100% and you'll get minimal time gains. I don't think you can ever get a 100% efficient use of traction without TC, based on the sensitivity and delicacy of the leg muscle movements.

Also, it would vary greatly according to track. In a fast corner, the difference on the exit of with/without TC is minimal, but from a slow corner, the gains are more. But then, the gains will increase again if the corner you've just exited leads to a long straight. eg. traction out of the Monaco hairpin will probably give less time gain compared to traction out of Lesmo 2.

[henrykleinespel]

I know it's not possible to precisely quantify that...
But maybe someone knows if it's more around 0.1s/lap or 0.5 or 1s/lap...

Also, maybe it would be easier to quantify the gain at the start (???).

[bettonracing]

I don't have the F1 data either but have a recommendation:

The F1 gains from traction control are likely to be a smaller percentage compared to an FSAE car. (Higher downforce, corners with larger radius, etc). As such, it seems reasonable that You would be able to better justify Your project (i.e. show a larger disparity between TC on & TC off) using wheel slip/ speed data from previous years, or beg/ borrow/ trade with other teams. You'll show larger potential gains, and will have a much more detailed analysis of where the gains come from.

While F1 may carry more glamour during the presentation, IMO having a detailed analysis of Your specific application would make it an easier sell. You can always throw in a picture of an F1 car at the end.

Of course, since I'm not the one approving your project, take this recommendation with a grain of salt.

Regards,

H. Kurt Betton

[Tim.Wright]

A simple justification without having to do a laptime simulation would be this;
It will (nearly) eliminate power on oversteer, which is a very desirable trait especially for an amateur driver who is searching for the limits.

Tim

[henrykleinespel]

FSAE seems like a very good idea! I'll try that! It should be much easier to get in touch with an FSAE team than F1

[mep]

Hi henrykleinespel
I already like that you think about ways to quantify the difference between traction control on and off.
Can't you try this out on a PC game?
I don't know if this is appropriate for your academic level but it's a very cheap and fast way to get some laptime differences. Dependant which programme you use you will also get some telemetry data.

Some other approach could be to find some reason why its very hard or impossible for a driver to accelerate always on grip limit. Maybe the pedal position is not rising linear (or cubic or whatever) because of changing conditions like increasing downforce, engine characteristics, gear changes, cornering, one wheel on curbs....

[henrykleinespel]

Well...my first approach was to build a numeric model of an accelerating car, taking into account parameters such as drag and lift coefficient, tyre grip, engine power, car mass, CG position, etc.

There ar still a few glitches here and there, but I'll soon finish this first approach. If anyone's interested, I can share it once finished. Currently, I get some good results, but the numerical values are wrong (something like 10-15s from 0 to 60mph for an F1-similar car).

[Tim.Wright]

How are you modelling the difference between a car with tcs and one without tcs?

[henrykleinespel]

The car with TCS (I use "TRC" ) is always accelerating with the optimal slip-ratio (and thus an optimal friction coefficient)...this allows me to calculate the force accelerating the car with TCS.

For the car without TCS, there are different options.
For now, I just did it so there is a constant, "non-optimal" friction coefficient when the slip-ratio is above a certain limit (simulating wheelspin and the associated traction loss).

From my first calculations, there is always a speed where the TCS becomes "useless"...so this only confirms what we already knew: TCS is useful at the beginning of an acceleration

Sure, I could do something more accurate...but this is only a numerical model and acceleration without TCS differs greatly from driver to driver...

[mep]

Well...my first approach was to build a numeric model of an accelerating car, taking into account parameters such as drag and lift coefficient, tyre grip, engine power, car mass, CG position, etc.

Just some weeks ago I did something like this with excel.
Check this even when its focused on overtaking:

viewtopic.php?f=6&t=9474&p=225117#p225117

I think it got lost here because I got no reply for it.
It might be tricky to model the difference between driver and TC acceleration so it might be easier to just show difficulties (non linearities) the driver has.

[henrykleinespel]

Looks really interesting! I just had a quick look at it and I will look further into it after I get some good sleep!

Right now I'm using an algebra software (Maple 14) but is all this works fine with Excel, I'll stop using Maple... (It's good with algebra...but bad with numerical calculations )

[mep]

I used excel because it is simple to use and it gives you nice graphs as output.
When the model becomes more and more complicated it might be better to use matlab (the only other programme I know). For me matlab is more complicated and I would need time to learn the programme first.

Something I should add. At the time I posted those graphs I did not take dynamic weight shift into account in the model. I didn't expect it to make a big difference. After I implemented it I was surprised that it really makes a big difference so I have different results now.

[henrykleinespel]

Yeah...I wanted to leave that weight shift aside too...but after a quick calculation, I realised that would not be ideal

[Sayshina]

As others have suggested, I would look at other formulae as getting accurate information, or even a response, out of F1 teams might prove dificult.

If you tried to use a video game however, the only thing you'd be comparing is the programmers idea of traction control, which I would assume would be a no go for an acedemic project.

If you want to find some F1 numbers to add a little depth to your project, you can look at the first year TC came in. Teams were using TC at some points and not at others, often during the same weekend.

McClaren used 1 year old Fords and had TC, and Senna of course. Benetton was the works Ford team, and with new engines had no TC, as Ford was aparently afraid it would destroy their motors. Oh, and they also had some guy named Schumacher. McClaren pretty reliably blew Benetton off that year.

Interestingly, somewhat counter to what's been mentioned here, in most formulae you find that the best drivers are actually faster without TC, or most of the other "driver aids". Usually, their single fastest time through a given corner will be with no aids. Their best 10 times however will be with the aids turned back on.