All you guys need is this program:

http://home.earthlink.net/~patglenn/ct.html

They guy already did all the work and calculations.... it's pretty accurate.

 

The numbers for the S2000 are wrong, though (take a look at the gear ratios). Has anyone done this with the real S2000 configuration?

 

[QUOTE]Originally posted by Tanqueray
[B]The numbers for the S2000 are wrong, though (take a look at the gear ratios).

 

Wouldn't be so quick to "DUH" someone. I think with getting some baseline dyno runs at the wheels of various cars you could modify the equations you proposed for some additional accuracy by taking into account the "peakiness" of the S2000 and trying to figure out the "Area underneath the torque curve" and "the rev ranges at which you will be operating in".

Strictly Newtonian . . . no quantum here

 

------------------------------
...i challenge you (or anyone else) to come up with a better method that is anywhere near as accurate AND simple. ...
------------------------------
I'm sorry but the fact is that this model pressented here is just as INNACURATE as a random-number generator. Sorry if you don't like it but all that this model does is really work out the "weight to power" ratio and then apply a scaling factor to it - that is all. Do you seriously propose that "weight to power" is an approximation of acceleration? Yes, more power and less weight the better but it will not give you any number that's usable in an argumant regarding acceleration. You might as well just use the "weight to power" figure and claim that it demonstates acceleration.

Maybe you should read my post again and find some helpfull info there - like the fact that it's the 'average power' in the power-band (given by gearing) that you need to look at. Without it this kind of equation is totally useless.

ps. Even a 'complex' erquation, accounting for all the facts I mentioned, will give you only a very loose approximation of what the acceleration will be. So what I'm proposing is already a 'loose approximation' - let alone the equation that you're talking about.

 

---
I'm sorry but the fact is that this model pressented here is just as INNACURATE as a random-number generator.
---

guys... "accurate AND simple". see that part that says "AND simple"? sure you can do some really fancy software modeling, but i really doubt you get as much estimation "bang for the buck" as with a simple back-of-the-envelope formula. sure it's not perfect, but then what equation is ever going to be perfect? if you only want to guess to within +/- 1/2 second or so, i think R&T's does fine. and if i'm not mistaken, that's what aaron was asking for.

---
Wouldn't be so quick to "DUH" someone.
---

the "duh" was not meant to say i thought he was wrong per se. it meant that of course you can always get a more accurate estimate with nitpicking and hairsplitting, but what's the point? the original post's challenge was to estimate 1/4 mile times for a number of different vehicles, so optimizing the formula for one specific car's torque curve would be useless.

I love how someone always has to pipe up with [Martin Prince voice] "owmmm, *actually*, that's not *entirely* accurate...simpletons!" [/Martin] These are the same people who will tell you that you can't use 3.14159 for pi because really you should use 3.14159265358979323. whatever.

 

-----------------------------
if you only want to guess to within +/- 1/2 second or so, i think R&T's does fine. and if i'm not mistaken, that's what aaron was asking for.
-----------------------------
That's the thing - this 'simple' algorithm is not accurate to within +/- 0.5sec - it's maybe accurate to +/-2secs (maybe even 3 secs). This algorithm is is only as informative as "weight to power" ratio - if you want simple then just stick with that as the car with a better weight to power ratio 'should' be quicker.

Though "weight to power" ratio does not always guarantee a faster car let alone "by how much" faster. Maybe the point I missed in making clear is that it is totally pointless using this formula and then debating that car X is faster than car Y by 0.5 sec because this furmula sais so. Same with cutting down the 1/4 mile times - this formula has almost no relevance on how much weight you need to lose (or power you need to gain) in order to improve the times.

It's fine having a 'simple' formula but you have to realize that this will tell at most who is faster - though even that is correct in maybe 70% of cases. It will never tell you the time ... well, it will tell you a time but it could be good 2 - 3 secs off what it really is.

 

Even with the formula quoted by dsp. Arent you guys forgotton something?
Power loss to the wheel??
The 240hp is only HP on the flywheel. You still have to take about 15% loss on RWD. The formula is just a good estimation on avg HP on the wheel w/o considering power loss.

 

---begin quote---
maybe 70% of cases. It will never tell you the time ... well, it will tell you a time but it could be good 2 - 3 secs off what it really is.
---end quote---

2-3 seconds?! yeah right. put your money where your mouth is: name a car for which the equation miscalculates by more than 2 seconds. i dare ya! oh yeah, make it a meaningful car. no automatics, hybrids, or vans or anything.

Here's numbers for a few cars with different engine & drive-wheel setups. First number is formula estimate, second is actual test results from various car mags. The estimates are pretty damn close if you ask me. Especially if you allow for the 5% HP variation and fat driver syndrome.

Ferrari 360 Spyder: 12.2/12.9 (mid engine RWD)
BMW M5: 12.8/13.2 (front engine RWD)
Porsche 911 Turbo: 11.7/12.3 (rear engine AWD turbo)
Chevrolet Corvette: 12.2/12.7 (front engine RWD)
1992 Nissan Sentra SE-R: 15.4/15.8 (front engine FWD)

---begin quote---
missed in making clear is that it is totally pointless using this formula and then debating that car X is faster than car Y by 0.5 sec because this furmula sais so. Same with cutting down the 1/4
---end quote---

i don't think anyone will disagree with that. pissing contests are stupid in general, let alone ones that are based on an obviously simplified method used to calculate a number of dubious significance in the first place (1/4 mile times). that's almost as dumb as nitpicking that frictional losses or torque curves or astrological alignment or the kitchen sink should also be included in the calculation.

the equation can be handy, though. maybe for seeing how much HP you "really" gained after some new mods, or whether you can beat that friend who you usually have by .25s but who got new cams & header.

 

----------------------------------------
-3 seconds?! yeah right. put your money where your mouth is: name a car for which the equation miscalculates by more than 2 seconds. i dare ya! oh yeah, make it a meaningful car. no automatics, hybrids, or vans or anything.
----------------------------------------
Ok here's a short list, theyr specs and average 1/4mile time:
- MR2 (pre Spyder, N/A) - 1230kg, 117kW (158hp), 1/4mile = ~15.2
- MR2 Spyder - 960kg, 104kW, 140hp, 1/4mile = ~15.5
- WRX (pre '01) - 1260kg, 155kW (210hp), 1/4mile = ~14.2
- 200SX S15 - 1260kg, 147kW (198hp), 1/4mile = ~14.7
- M Roadster (US spec) 3070lb, 240hp, 1/4mile = ~13.8
- Lotus Exige, 795kg, 133kW (180hp), 1/4mile = ~14.0
- TT Roadster, 1550kg, 165kW (222hp), 1/4mile = ~15.2

Though, you're right, there will not be 2-3secs discreptancy using the 'simple' formula .... I was commenting on 0-60mph times - they are a lot more varied than the 1/4mile times. Though, still, even using the 1/4mile times some of these will be over 1sec out .... 1sec is a big difference in the 1/4mile times. Cars that are 1sec appart on 1/4mile are in very different leagues.

If you could get the 'average power' instead of 'peak power' then I bet the numbers would be a lot more relevant.