Excel / math help needed -virtual volvo 300 link

[jtbo]

Even this is someway Volvo related, virtual Volvo that is

I have now bit of problem and I would need to make a graph in excel, it alone is not a problem, but making data for graph is an issue.

I have start value which is 0 or 1 depending from source, then I have final value which is variable, but lets say that is 1.41 (y-axis), then there is variable, 12000, which should set scale of X axis and final value should be at that 12000, might be a lot higher too.

Then there is initial slope parameter, for example -3,5E-6 that if I have understood correctly changes curvature of graph and changes how to end to that 1.41 which is at 12000 of x axis.

Now which kind of formula do I need to get numbers for example from 0 to 1.41 for 12000 (or 1200 if we divide by 10), so that initial slope variable will change how numbers increase and decrease, so that I can make graph from those numbers?

All my attempts seem to result need of making new graph each time as I change 12000 to something else, that is I manually have to copy formulas over that many lines and I really think there should be much easier way, maybe a macro or something?

Sorry if it sounds confusing, it is quite confusing for me as it has been way too long since I had much to do with such stuff.

[Chris_C]

If you have both those values in a pair of columns Jani, you can just plot x vs y on the data, that's how I did the ones for my knock graphs. Like this? ( http://farm7.static.flickr.com/6204/611 ... 6629_o.png ) but your data should be linear and make a rough line rather than being all over the place.

If you have too much issue, chuck the file on your hosting and I'll have a look

[jtbo]

It seems to be that my initial thinking was bit off so that problem actually don't need solving, however I ended up to another problem of same subject instead.

I need to solve a and b:
a*t^3 + b*t^2 + c*t + 1.0

that should give 0,3 as a result to here, but of course not 0,3 but some other number as 0,3 is wrong value, but it should give correct value to that LSFM parameter:

Code: Select all

c          LSFM       LSFL        t
-3,50E-06 0,300000     12000     5589,2475

There is also 284,875 that is one front tire actually and that is t which is same in Newtons.

This one then gives final multiplier used in other calculations, which gives me result of 0,8985628312928 with parameters above:
((2 + LSFL*LSIS - 2*LSFM) / LSFL^3)* Fz^3 + ((3 * LSFM - 3 - 2 * LSFL * LSIS) / LSFL^2) * Fz^2 + LSIS*Fz + 1

0,3 is wrong value and I try to find correct value and that a and b thing should give it, but how do you solve a and b, is it possible with this data?

Here is the file:
http://jtbo.pp.fi/tiedostot/rfactor/loa ... y_v01a.zip

3rd post in here has these equations and more information about it, however it should be simple mathematical thing if one is just enough good in math, or enough recently been at school that is :
http://www.racesimcentral.com/forum/sho ... ng-the-car

[Chris_C]

I find it amusing when I google for some background to this I find threads with you and Alex Laidlaw talking about drift mods

I'm having issues with it atm, my maths isn't hugely stunning but passable on a good day.

Grip multiplier = a*t^3 + b*t^2 + c*t + d

where t = current load as a fraction of final load, or (current load / final load).

Any load greater than the "final load" returns a grip multiplier of "final grip".

In that equation, d is 1.0 and c is the initial slope, so you really only need to solve for a and b.

The form of that equation used by ISI is

mult = ((2 + LSFL*LSIS - 2*LSFM) / LSFL^3)* Fz^3 + ((3 * LSFM - 3 - 2 * LSFL * LSIS) / LSFL^2) * Fz^2 + LSIS*Fz + 1

where:
LSIS = LoadSensInitialSlope - first parameter
LSFM = LoadSensFinalMultiplier - second parameter
LSFL = LoadSensFinalLoad - third parameter
Fz = NormalLoad on the tyre - in Newtons

Now according to that, those two equations are completely independant? As for t to be (current load / final load) it cannot be Fz as the second equation shows?
So, my question would be, the program calculates "multiplier"? If so we are trying to define the 3 values (LSIS, LSFM, LSFL) for each LoadSensLat and LoadSensLong? Which as you define in your xls is different for front and rear tyres?

[jtbo]

Well, I have been working more with it and according one that knows LSFM would not be possible to calculate, but should be just tried and experiment, but on the other hand real data should be used, so I'm bit confused.

Anyway here is more recent version showing bit of effect:
http://jtbo.pp.fi/tiedostot/rfactor/loa ... y_v01a.zip

edit: program uses loadsens parameters all together to get multiplier for current load of tire.

I'm still having bit different result that what one program gives, but I suspect it is because graph is smoothed to end value or something like that, don't really know.



If I wanted to end of graph be LSFM value and not cut to it but smoothly joined to that, I guess there would be function that would adjust curve to meet end value neatly?

[Chris_C]

If it was Matlab it'd be a case of

Code: Select all

if (x>LSFM)
   {
   x = LSFMval
   }
else
   {
   x = reallycomplicatedformulathatmakestheabovecurve
   }

I imagine excel can do similar!

[jtbo]

Hmm, that would be:

Code: Select all

IF(x>LFSM;reallycomplicatedformulathatmakestheabovecurve;LFSM)

However I'm already using that, it makes that big drop at near 21898, I remember reading there should be some parameters that can affect greatly how data is plotted, but so far I have stumbled upon a trendline and that does not quite cut it.

[jtbo]

Ah, there was only one problem, in that monster formula, first D11 was C11, after fixing that it works just right with my IF clause, wooden eyes, can't see anything

[Chris_C]

Nice one dude

So, what tyres are you putting on the beastie? I take it this is for the 340 and not Jani's second awesome project (tm) ?

[jtbo]

Well, currently 185/60-14 budget 'sport' tyres which I happen to still have under the car, but others will of course follow.

There is actually link for download if one happen to own rFactor that is:
http://www.rfactorcentral.com/detail.cfm?ID=Volvo%20300

What comes to 2nd awesome project, well, I have been making cars to behave for this one, Brock Yates is making cars look nice, I think he is doing it very well:
http://www.nogripracing.com/forum/showp ... stcount=21