Could a formula work?

[Eagle7]

I'm getting pretty weary with all the hubbub about which stock HP to use when classing a car, and what subjective evaluation to apply to that HP to properly class the car. It got me thinking again - could a formula work?

I'm an engineer, and if I've learned anything in the last 40 years it's that physics is pretty darn predictable. I don't know much about what does and what doesn't make HP, but there are a lot of you out there that do. I'm totally convinced that given enough information about the design of an engine, a good engineer could calcuate its potential HP in IT trim within a pretty tight tolerance - maybe 1 or 2 percent? If that's the case, it seems like a much better approach than guessing how much gain a particular model would achieve over stock HP.

If you could set a formula for engine HP, I'm confident you could also do it for all the other adders and subtractors. Ideally we'd end up with a published formula that anyone could use to calculate the weight of any car of interest.

So, you gearheads and mechanical engineers out there, how say you?
Is it possible to accurately calculate the HP potential of an engine from publicly known design information? If not, can you cite examples?
Is it feasible to perform this calculation for the full set of IT cars?
Is it feasible to calculate the other adders and subtractors of interest?
If all this is feasible, would it be a better system than using stock published HP?

[GKR_17]

Short answer no.

A formula was proposed a few years ago, and even though there was a significant amount of work, it still had enormous holes. There's just no way to get enough acurate data for every car. Heck, look at SM, they only have 2 chassis options, and 3 basic engines to work with and they still debate the spec weights.

The process does seem to be a big improvement over the old closed door system, and then we fall back on the non-guarantee of performance. I sugest you develop your own formula to pick the right car and be glad for any advantage you were able to find. Otherwise you might think about a spec racer.

Grafton
(mechanical engineer)

[Knestis]

Other answer, is "yes."

I've got spreadsheets of data going back to before 2000, playing with this question. I agree with the proposition that it could get close - given enough data - but that (1) there are a lot of factors, and (2) the real problem is finding the raw data.

I actually went so far at one point as to start playing with statistical analysis software (SPSS) and regression formulas, to see what factors were the best predictors of performance. However, I quickly came to understand another problem: that the outcome measure I was using (lap time?) was influenced by the DRIVER, perhaps even moreso than attributes of the car, so it became a very difficult proposition to do the math.

It might be more sensible to tackle it as a pure physics problem. The physics engines of modern racing video games and lap simulations actually do a pretty good job (from what I'm told) and ultimately, there really are only so many variables in the CAR.

Volumetric efficiency of the engine - bore, stroke, rpm, valve and intakc cross-section areas. It's all about moving gas and turning chemical energy into heat energy, into kinetic energy.

Brakes turn kinetic energy back into heat energy. If you ignore all of the exogenous variables (ambient air temp, etc.) which can be held constant since we're just comparing designs, it should be possible for the laws of thermodynamics to predict how effective brakes will be.

Something like 80% of the energy in fuel gets used to push a car through the air at highway speeds, and aero drag increases with the square of velocity, so an ITR Bimmer is using a lot of horsepower when it's at the end of the straight. Since the planforms, parasitic drag qualities, and other vague factors are going to be similar for all IT cars (since they pretty much all follow the same basic design parameters) aero drag can be pretty accurately accounte for considering Cd and frontal area. Or as Greg suggested elsewhere, coast-down data is pretty repeatable and reasonably accurate if all we care about is the "negative horsepower" that the atmosphere imposes on us at speed.

Weight is a variable we can control to a degree. Mr. Newton can tell us what a given car is doing in terms of accelerations at any given point on any given track. The problem becomes a big calculus mess since, to get accurate numbers, we have to consider the math at a bazillion points around a huge number of tracks. And come to some consensus about a methodology to define a "mean" track, that represents a compromise of all of the real-world tracks we need to take into consideration.

Etc.

THEN all you have to do is figure out how the variables interact and come up with an equation into which you can plug the values.

Simple, right? :)

In the real world, we kind of go back to the regression formula idea. There are some factors that make a large difference (torque multiplication due to gearbox ratios?) and some that influence lap times less (CG height?). It should be possible to account for a large percentage of the theoretical sum that defines "performance," using some limited set of variables. Some will by necessity be left up to the entrant (e.g., tire size = drag, increased traction, and aero drag), since the rules allow variation there, so they can be set free.

K

EDIT - I'd propose that SM has its difficulties because the adjustments are made based ultimately on on-track performance data, confounded by driver skills, individual perceptions, group-think, etc.

[Lael Cleland]

I am not an engeenier, and i cant spell to save my life, but i have built engines, used a dyno, and blown up alot of engines. the prob i see about a formula, for potential HP is any thing form valve seat angles/ size of seat,cam timing,rings abuility st seal, and the friction it makes when they seal, controling windage, loss of power due to viscosity of the oil, even casting of heads being slightly diffrent. As individuals, they dont amount amoun to much, but all together could be substantial.. The only engines i have ever built are watercooled vws, there is just to many inconsistancies, in cast iron/aluminieum IMOP....

I repeat im not an engineer, but wish i was :D

[lateapex911]

Is it possible??

sure....

Is it practical??

Doubtful ..........esp for a resource limited club.