drivetrain loss in relation to rwhp?
This is why I can't accept the 'fixed horsepower loss' theory!
I see two discussions going on here, so let me clarify what you are debating.
Inertia.
Take that torque wrench and whip it around 180 degrees within one second.
The needle will deflect more than it would if the torque wrench was moved
the same 180 degrees within an hour period.
There is an opposition occuring which is clearly seen when the rate of
twisting the nut is increased. This is your analogy to the higher HP motor
wanting to move the car from a dead stop and accelerate it faster.
This is why I can't accept the 'fixed horsepower loss' theory!

Pull the engine and dyno it at the crank
Than put the engine back in the car and dyno to the wheels.
Subtract, divide.......voila.
As for A to B testing here is what we've found down under. See, down there, many of the shops have Engine dynos and Chassis dynos. The dynos are made by the same company, Dyno Dynamics.
The Dyno Dynamics engine dyno is nice as it is fully OBDII compliant so you can monitor your PCM while you dyno.
Anyhow, in speaking with Peter, they have done A to B testing up to around 730 RWHP. The loss through their cars is right at 110 HP regardless of HP. This isn't with 1 or 2 pulls we are talking about. I think Peter has had 8 or more different engines including H/C and 422 strokers. N/A, S/C, and single turbo setups. In one marathon push they did 8 or more different sets of heads and 10 or 12 different sets of headers. This ammounted to over 300 dyno pulls on one engine. After which it was installed into a 3900lb Ute and went 11.4 @ 120 MPH in 2002-2003.
The Dyno's used in your example at Sam's Performance were in fact Dynolog Engine/Chassis Dyno's, not Dyno Dynamics, there is a quite a difference between the 2.
In my fifteen years of by choice, by profession, full time, operating chassis dynos mainly within the high performance/race car applications, it has been my observation that drivetrain losses are more of a percentage factor than a fixed HP number for either low or high HP vehicles. This is because I have consistantly come very close in estimating a vehicle's factory advertised FWHP/TQ simply by applying a given percentage for the different types of drivetrains such as car or truck, manual or auto for low or high HP applications.
In my opinion this could be true because of the RPM componant of our drivetrain losses. Because we operate and test with increasing RPM sweeps, as LS_RX-7 stated in this thread on Pg 1, item 2, "a greater than linear power loss with RPM", the RPM losses dramatically offset the many other losses, some increasing, some decreasing in percentage as different loads are applied.


