Bingo, I think we might have a winner. I would love for some engineering geek to disprove that but it sure seems to make sense that drivetrain loss should be a fixed number on a particular car regardless of the crank HP. If you were to build a 996tt motor to put out 1000 hp, which has been done, would drive train loss be 200hp, or half the output of a stock motor assuming 20% AWD loss? I think not.....

 

Each drivetrain loss formula would be different not just because of the obvious losses of different computer software on the dyno,,differences in dyno,weather/altitude[correction factors arent that precise] and drive train condition BUT also Because the vehicles tire, wheel and brake rotor weights which are not a set standard either..

 

Because losses aren't a fixed number, they're a percentage.

Different example, but if you detuned a 996TT down to 100 bhp, would you expect it to only make 23 whp? I wouldn't ...

I have seen some good articles on this subject - Modified Magazine did one, albeit on a JDM. I think the conclusion was, crank to wheel loss percentages weren't static as power increases, but the horsepower/torque consumed by the drivetrain isn't fixed either, so the net result was... don't bother trying to extrapolate crank hp from wheel hp on a modified vehicle - if you really just "must-know" bhp, you have to pull the motor and put it on an engine dyno, at every power/mod level... it cannot be figured out with any meaningful accuracy with a chassis dyno reading.

 

I was saying that drivetrain loss should be a fixed constant on a particular car with all things equal except a varying HP output. In other words, if drivetrain loss is 80hp on a 420bhp car, and you add a flash, exhaust, intake, or whatever to the same car then drivetrain loss should still remain at 80hp even though the car may produce 550bhp for example.

 

I was saying that drivetrain loss should be a fixed constant on a particular car with all things equal except a varying HP output. In other words, if drivetrain loss is 80hp on a 420bhp car, and you add a flash, exhaust, intake, or whatever to the same car then drivetrain loss should still remain at 80hp even though the car may produce 550bhp for example. I am not sure if this would still apply at the hi and low extremes however.

 

The answer is the change in hp does nothing. USING the hp changes everything. With more speed there is more resistance or drag. Drivetrain components provide frictional/rotational drag. If you try to push 415 hp through it at once, you will have less resistance (because parts have to accelerate and move at lower speeds) than if you were to push 600 hp through it.

I would be willing to debate that the percentage is a not constant percentage across the whole powerband, but to say that the same drivetrain produces the exact same resistance when trying to accelerate it with 100 hp vs 600 hp, I would not be willing to do. It takes more power to get anything moving faster.

There are also plenty of resistance vs speed calculations that add up to real world performance with the use of a constant figure as lost hp. See hp calculators from trap speeds and such. You don't come close with a constant number

 

[QUOTE=c32AMG-DTM;3664520]Because losses aren't a fixed number, they're a percentage.

Different example, but if you detuned a 996TT down to 100 bhp, would you expect it to only make 23 whp? I wouldn't ...

I have seen some good articles on this subject - Modified Magazine did one, albeit on a JDM. I think the conclusion was, crank to wheel loss percentages weren't static as power increases, but the horsepower/torque consumed by the drivetrain isn't fixed either, so the net result was... don't bother trying to extrapolate crank hp from wheel hp on a modified vehicle - if you really just "must-know" bhp, you have to pull the motor and put it on an engine dyno, at every power/mod level... it cannot be figured out with any meaningful accuracy with a chassis dyno reading.[/QUOTE)


I would think for each car, since it would vary from car to car, you would have to do a chassis dyno with monitored factors And then pull the motor, lol, than put it on an engine dyno with the same factors...That would create that specific cars drive train loss..Then you would have a baseline for your future mods to estimate crank HP with that ratio.
I wish , as everyone else, that there would be a easier way to get it persice

 

That makes sense. The real number is not fixed nor is it a fixed percentage either but something in between it would seem.

 

You have flywheel horsepower which is the most accurate because there is no drivetrain loss, no traction variations that can be caused by different tire size, width, diameter, friction depending on the compound of tires(I have been told slicks rob more power because they are stickier), along with wheel weight, tire weight, flywheel weight, clutch weight or torque converter weight, even driveshaft and\or axle weight; all these things come into play when you strap a car on a rolling chassis dyno and that is why the numbers are less than what they would be at the flywheel. Altitude, humidity, ambient air temp also changes things. I like the idea of Dyna Pack dynos because by bolting on the units directly onto the hubs, it takes some of the above factors out of the equation. I am sure you have seen those dyno videos where they have 5 fat guys in the trunk of a car while they are trying to dyno a car because they need the weight over the rear wheels for traction. With Dyna Pack dynos you have 100% traction unless your clutch or torque converter is slipping.
Ideally you would take your engine out of your car and take it to an Engine Dyno to get the most accurate number but most of us will not do that; takes too much time, costs too much money.
I don't get this whole fake hp numbers, etc. stuff that is being thrown around.
I would rather call it CLAIMED or ESTIMATED HP versus MEASURED\WHEELHORSEPOWER.
Even Wheel HP will vary from dyno to dyno! Mustang dynos are known to show lower #s.
Someone earlier said something like he estimated the engine would make 490 at the flywheel but later said it would be a higher number at the wheels which is completely wrong!! I am sure he meant it the other way around or made a mistake putting in the numbers.
I agree with getting a baseline and then using the dyno to see the improvement after the mods.

 

I would argue that WHP is the most accurate, it doesn't matter how much hp you have at the crank if it isn't there at the wheels. WHP is what moves the car.

Also, the HP loss at higher torque is accurate because the gears are pushed together harder causing more friction than when less torque is being applied. This is the same for bearings also. I will agree that the percentage should go down as the HP figures go up because there will be some losses that are constant, such as the power it takes to turn the gears through the fluid (engine and drivetrain) as long as the RPM is the same.

 

FWIW one of our local tuners told me they made over 560 whp on a 996 GT2 on k24's with just tune exhaust. Now thats one I will have to see to beleive. Good thing is that I see the car at the track so if im ever around it on a straight ill know for sure.

 

A18g has a hard time let alone a k24 lol

 

If that were true, a lot of us, including me, have wasted a lot of money!

 

I agree. Plus K24's are pretty flaky turbos anyway and with so much more to be gained with the simple (K24/18g or 20g) to the exoctic Alpha 28's turbo swap. Just do the turbo swap and get less lag and way more power. IMHO

 

True. Nice to see ya Marek !

MK