This can be done on all rolling raods, which includes the Dyno Dynamics one despite the claims of some operators (see article in Redline Feb. 2009 page 21).
Ambient/intake air temperature probe
Every 3 degrees Celsius inlet air temprature increase decreases the by around 1% or 10 degrees by nearly 4%.
This is due to fact that the volumetric weight decreases with increasing ambient temperature. If the temperature where the inlet is located is deliberatedly high an engine with a specific power output can loose a substantial amount of power.
If for example the temperature probe is placed in a very warm environment in of about 20C higher than the actual ambient temperature on an engine with a specific power output of 190bhp the print out would show anything between 205 and 210bhp.
If the very same engine would be operated in a very confined space without the heat dissapating even blowing air onto the engine would not give correct results if the probe is in a much cooler spot. The results can be as low as 175bhp at 20C lower probe temp reading.
Wheel alignment
If the wheels aren't properly aligned the reading on the graph will not be a true reading. This can give wrong results of in excess of even 10%
The operator can influence this by ignorance as well as creating varying results when changing the wheel angle in relation to the rolling road wheels when doing several runs.
Applying brakes
When applying brakes on the run up some of the wheel power will be absorbed by the brakes. The same applies when transmission losses are measured on the run down.
On the run up it would show lower wheel power than the real figures and on run down it would add up to the losses, which would give an inflated flywheel figure.
Throttle position on power runs
Any power run will be different on full throttle or part throttle. Generally the full throttle run will provide a graph with more power than a part throttle run or a run with varying throttle position.
Altering transmission loss data input
This can be done on the Dyno Dynamics too (see above mentioned Redline article). By giving in different runs different transmission loss codes the corrected flywheel graph actually becomes incorrect.
Loading the rollers
Different loads on the rollers will affect the printout on the graphs as well. This can be done in a number of ways (which I am not going into as it is possible to do this in a number of ways), which can be intentional as well as ignorance.
Other factors
The following factors can influence the readings too but are difficult to create intentionally before the audience on a rolling road day.
Tyre pressure differences.
Tyre profiles.
Tyre compounds.
Tyre temperature.
Engine temperature.
Tracking. Especially towing in or towing out.
Transmission fluids. Especially when their properties aredifferent then the ones normally used.
State of gearbox/Diff.
State of clutch.
Brake bind.
Malfunctioning Lambda sensor(s)/MAPsensor(S). Both types of sensors when not in proper working condition can vary results dramatically, which can be from run to run different but also consistently show poor results.
Other factors that can affect the runs especially when the runs are over a longer period apart
Exhaust condition
Air filter condition
Fuel differences
Variations on different rolling roads.
Different operators.
Maintenance.