some info here:
https://www.evolutionm.net/forums/ev...e-acd-ayc.html

 

There's little to no relationship between drivetrain loss and engine power.
It takes less than 40 HP to push a car down the road at 70 MPH and it takes about 100 HP to push a car along at about 100 MPH.

Drivetrain loss on a dyno is a fraction of those above numbers as there's no air to push out of the way on a dyno. Just the wheels and rollers and tyres to spin with the associated gears.
The single biggest factor is the tyres.

Drivetrain loss increases with road speed.
So therefore if you don't factor in the road speed, you will never know the loss.

 

that is just false... drivetrain loss has several elements.. some are speed related (like oil drag loss) but a pair of gears has friction losses that are directly proportional to power they transmit. Some rolling roads try to measure drivetrain loss by measuring deceleration but that is unloaded and drivetrain loss will not be the same as with the power on.

Yes, it is wrong to just say that evo drivetrain has flat 22% loss... but power loss trough the drivetrain does depend on the power.
The measurements made in the UK, with same engine being run on a rolling road dyno and a proper engine dyno yielded that 22% loss for that specific power, so it is a good rule of thumb measurement.

 

Some rolling roads are calibrated in such a way that, if for example you put a car on it that makes 300HP at the flywheel, it prints out 300HP but reads it at the wheels.


It's nothing more than an inbuilt calibration which tries to account for an unknown drivetrain variable by treating it as a fixed % when it's not, never has been and never will be.


The only truly accurate dyno is an engine dyno.
One that's calibrated to do emissions (J1349)
Drivetrain loss = zero.