The argument of power measurements at the wheels versus at the crank opens a huge can of worms whenever it is discussed!
You've hit one of my soapbox subjects now so I can only appologise for what is about to follow
![Roll Eyes ::)](http://images.omegaowners.com/forum/smf2000/Smileys/oofstd/rolleyes.gif)
.
The truth is that 200 BHP dissipated in the transmission would kill it very quickly.
The most accurate way to measure an engine's output is to bolt it into an engine dyno which measures the power directly at the flywheel with the engine out of the car. Few people bother because of the cost involved and the time taken to install an engine in a test cell, measure it and then remove it and clean up.
More commonly a rolling road is used because the engine can be tested in the car with only a few minutes of setup time (drive the car onto the rollers and strap it down). You will not get a true figure of engine output by doing this due to losses in the transmission and, more importantly, by the tyres driving the rollers of the rolling road. This isn't such an issue if you just want to make a relative measurement, or to tune the engine for best power because you will be able to see even the slightest relative change in power, but making an accurate absolute measurement is more tricky.
The problem is, people want to go away from a rolling road with a figure to brag about down the pub and that means trying to work back to a figure at the flywheel because that's what manufacturers publish. This can be achieved quite accurately by doing a power run in (normally) 4th gear on a manual (because it's usually the direct gear and therefore the least lossy) up to and normally a little beyond peak power RPM, measuring the torque and speed at the rollers (hence power). The test is then carried out in reverse by declutching and measuring the deceleration of the rollers, wheels and transmission as they slow from the peak RPM speed (which is going to be around 120 MPH or more
![Shocked :o](http://images.omegaowners.com/forum/smf2000/Smileys/oofstd/shocked.gif)
). This gives the machine a picture of the losses in the system as the whole thing slows down again, so at any RPM you end up knowing the power at the rollers and the losses between power at the rollers and power at the flywheel. You then need to compensate for the air temperature, humidity and pressure, as this affects engine output on any given day and you have a figure that, if measured carefully, will be within 5% of an engine dyno's figure or manufacturer's claimed output, if that's accurate.
The problem is, if this process is not carried out accurately the result is meaningless. It's meaningless on an automatic car anyway, because the gearing between engine and wheels is constantly changing as the torque converter does its' job and the losses are much greater when the power is on than when the transmission is coasting.
In addition, some rolling roads don't bother with a "coastdown" measurement and just add a completely arbitrary estimate of loss to the measurements at the rollers, often designed to tell the punter what he wants to hear about his engine! They often don't get their rollers calibrated regularly so the basic measurement is not necessarily accurate in the first place!
Don't make the mistake of thinking that power "at the wheels" is more representative of the true power available to you as you drive the car, either. The losses measured here are far greater than the losses experienced when the tyres drive against the ground because on a rolling road the tyres have 2 contact points with the rollers instead of just one against the ground, and the vehicle is often strapped down, putting much more load on the tyre - roller interface than would be present with just the vehicle's weight. Tyre pressure also has a big effect on these losses, and is frequently adjusted when a car is put on the rollers.
In summary, ignore the figure at the wheels completely and hope that the figure at the flywheel was measured with a proper coastdown test. The actual loss through the transmission itself will be much less than 200 BHP. A few tens of BHP at the most, I guess. Most of the losses come from friction between the tyres and rollers, and in distortion of the tyre as it runs on the rollers. They result in heat in the rollers, tyres and the air passing over them, which is why the transmission oil doesn't boil!
Kevin