legacycentral bbs

This has been in the back of my mind ever since I heard about it. Why does the drivetrain take a percentage of power to move, instead of just taking an amount of power to move? It just doesn't make sense to me that as your power increases you still have a percentage that is getting taken away by the drivetrain. Does it not make more sense that it'd take a certain amount of power to turn the inputshaft, gears, axles, etc instead of them taking a certain amount of power no matter what the situation? Let's just take some random numbers, let's say it takes 30 hp to move the entire drivetrain. So if you've got an NA legacy with 130 hp, you should be making 100 hp to the wheels. If you've got a turbo legacy with 160 hp you should be makeing 130 hp to the wheels. If you've got a WRX with 227 hp you should be making 197 hp to the wheels. Instead of it taking 30% or whatever where the respective numbers would be like 130chp/91whp, 160chp/112whp, and 227chp/159whp or whatever. Does anybody understand my reasoning? Why does it take a certain percentage? And how do we know it takes a percentage or power? I often see people saying they have x whp so they have x chp based on the percentage, but who said it takes a percentage?

It doesn't take a percentage or a fixed amount. Neither is correct. They're both approximations. You're absolutely right that people oversimplify it, and the resulting errors can be large in magnitude.

Hmm, so what then? How does it work?

I don't know if there are any good theoretical models of it... I think you'd have to measure crank power and wheel power at various engine and road speeds and then make a graph of it for any particular engine.

I think the big thing to take into consideration is that the amount of power loss in the drivetrain is probably not constant throughout the engine rpm range, as well as at different power levels (ie adding a larger turbo, etc)

As Vikash mentioned, I don't think there are any good models out there that represent this.

About the closest you could do is dyno an engine alone with as close to the same exhaust performance as it will have on the car, and then dyno the car as a whole. Then compare the curves, and see if your losses are fairly consistant throughout the rpm range, or if you have peaks and valleys where parasitic losses change with rpm.

Tire friction increases with speed also and I think it varies by gear and throughout the range. The percentage thing is probably people trying to define their guesstimates IMO.

Steve

Power lost to heat and friction vary with speed, so that explains why it isn’t a fixed number. Leaving the percentage number, which after thinking about it could vary substantially depending on any number of things. And in an almost related way I just read in the latest Motor Trend that a top fuel dragster’s exhaust alone creates 800lbs of downforce, just had to share.

wow, maybe I should turn my exhaust so it's pointing straight up, though with it pointing straight back it must be pushing my car forward like a jet engine!

I think you guys are going in the right direction here. At lower power levels, it appears to be a percentage, but as power goes up, that percentage drops. Which suggests some sort of absolute power loss through the tranny. Gearing varies this, as well, which is why a 2nd gear dyno run can be significantly different than a 3rd gear dyno run in power to the wheels.

Based on my conversations about this with Xephyr (who keeps immaculate notes, has every dyno graph from the last 4 years, and who models everything) he has come to the conclusion that the tranny in his RS losses right around 60-70hp. It seemed to be less at lower power levels, but once he hit a power range of around 250WHP, the losses got pretty consistent. This assumes that his modelling of what he is making at the motor holds true. But when I see him model in advance within 5hp what his new turbo set up is going to make once he gets on the dyno, I tend to trust his calculations.

the friction elements will tend to go up with RPM, but not likely linearly unless you begin to trash something. Friction definitely goes up with power (more torque provides a higher normal force on gear teeth, for instance, which gets multiplied by the coef. of friction), in fact nearly linearly.

The shear friction of fluids should be similar for whatever power you are running through the same drivetrain, so no matter the power of the motor these types of losses should be similar.

Inertial power losses should also be consistent regardless of the engine power, as long as the driven componenets are the same.

The biggest losses are from changing the direction of the power being transmitted; the more times you "bend" the power around corners, through differentials, CV joints, etc, the more power lost. This type of power loss is due to all three factors above, but it is the friction component that rises the most with increased power.

Simplifications like the 20% rule come about since many drivetrains are reasonably similar, and there is no good way to measure engine power directly, that is with said engine still bolted into the car. If you look at reliable manuf. dyno runs versus good tuner's rear wheel results (the "before" in tuning exercises), the 20% rule is pretty damn close most of the time.

Hmm, good discussion.
On a side note, what exactly is horsepower anyway?
Is it based on accelleration?

Theres a guy here in Hillsborough who builds Porsche race motors. Hes got his own motor Dyno, and says calculating Horsepower is a waste of time and its just an estimate anyway, where as you can truely measure torque.

Horsepower is torque multiplied by time, so if you could measure one accurately, then the other should be accurate as well. The measurement of HP has changed over the years, though.

Steve

Actually, horsepower is related to rpm, not time

The formula for hp is

Horsepower = (Torque * RPM) / 5252

5252 is a constant.

That's what I meant by time. Sorry, that wasn't very clear to those who don't know.

Steve

Revolutions Per Minute

Torque divided by time.

Another question.

How does one calibrate their Dyno?
I mean, when someone posts their Dyno readouts, how accurate is it really? I understand the purpose of a dyno and its really just a reference of how much power one has gained throughout the modification process, and for tuning. But if I wanted to impress everyone with a dyno printout, Id probably throw on the smallest tires I could fit?

the circumference of the tire is figured in. Not only that, but the measured torque with small tires would be lower