There really is some huge misunderstandings regarding Chassis Dynamometers!
Firstly, torque is purely a measured value - it is the torque that the dynamometer measures or, more accurately with a roller drum type dyno it is the calculated figure that is being applied to that drum to make it accelerate at the speed it is.
Power is a calculated figure - it is calculated from the torque and RPM (speed of the engine) - HP is the rate of doing work!
Gears are torque multipliers - if your final drive ratio is 3.92 and you test in 4th gear which is a 1:1 ratio then the correctly measured wheel torque will be 3.92 x flywheel torque (this does not take into account any losses).
Hence, the torque applied to the wheels is much greater in 1st gear than 5th gear but most dyno software divides this torque figure by the gear ratios before it 'spits' its figure out as we are not used to seeing torque figures of 800+ lbs/ft!
Urban myth No.1 = dyno plots showing torque and power should cross at 5252
The above statement is not true if the plot is of 'corrected HP'! As discussed the HP figure is derived from the torque:
HP = Torque x RPM ÷ 5252
However, so Dyno figures around the world can be compared a standard for temperature, pressure and sometimes humidity is introduced (SAE, DIN, STP etc) but because torque is the measured figure it shouldn't be applied to those numbers, only the calculated numbers of HP. Most Dyno software 'incorrectly' applies it to the torque figures as well to prevent the 'public' going up in arms about incorrect Dyno plots not crossing at 5252 rpm. (Mathematicians know that correction factors should only be applied to calculated numbers not measured numbers.)
So the actual equation for corrected HP is:
Corrected HP = (Torque x RPM ÷ 5252) x Correction Factor
Manufacturers quoted HP figures and your car on a chassis Dyno:
The measurements a chassis dynamometer makes are completely different to the way an engine dynamometer makes them.
When the engine flywheel power has been quoted by a manufacturer this has been obtained from the engine on a test bed, with an optimized exhaust system in place (no CAT), the induction system is completely different to what is in the car and all the fluids are maintained at ideal operating temperatures in an environmentally controlled test cell.
To highlight the above - most manufacturers fit the same engine to a number of model lines yet the quoted output of that engine will remain the same across the model lines even though the intake/exhaust systems are completely different and quite often the cooling systems are laid out differently as well.
Further to the above the test is carried out completely differently - the chassis dynamometer normally uses a sweep test (power run) to plot the torque/HP curves, during this run a significant percentage of the engine power is used to overcome the inertia of all parts within the system (internal engine parts/flywheel/clutch/gear sets/prop shaft/drive shafts and quite often wheels).
The engine dynamometer measures the power using what is called a step test - the engine is held in steady state against an engine brake (hence brake horse power) and measures the power at the flywheel. Since the engine and none of it's internal parts are accelerating there is no power being used up to overcome inertia so the readings are much higher.
So,
In summary – if you expect an engine to make the same power on a chassis dynamometer as the vehicle manufacturer or engine builder suggests (after testing on an engine dyno) when it is installed in your car with sub optimal exhaust systems etc and all the inertia it has to drive then quite frankly you are delusional!
Drive Train Losses:
Probably the loosest term used in the world of chassis dynamometers! The truth of the matter is modern transmissions are extremely efficient and if your gearbox/diff was absorbing any more than 10% of the power from the engine then you would, as Hammy states, boil fluids and melt gear cases in next to no time. Depending on the Dyno operator and the software the drivetrain losses can account for all manner of things including tie down tension, wheel deflection, rolling resistance eyc……the lsit goes on! It must be appreciated that no matter how these are derived they are guesswork and guesswork alone.
As a caveat to the above, as long as the guesswork is accurate and consistent then it really doesn’t matter – what is important is the consistency of the guesswork – as long as the operator is trustworthy then meaningful figures can be obtained from it and correlated to the manufactures output. A chassis Dyno is a tuning tool – nothing more. It is used to map engines whilst in their native home and, if you use the same one with the same operator you can see if modifications you have made improve the performance.
In Steve’s case he has taken his car to a rolling road and been given some figures – the Dyno operator who uses this ‘tool’ day in day out has then given him an ‘estimate’ of what the flywheel power would be if the engine was on an engine dyno. He is the only person who can do this so I really can’t understand people questioning it?
The reason the test was carried out in third gear might well be the fact that in 4th gear the speed may have been to high for the rolling road.
You simply can not say what percentage drive train losses should be on his dynamometer – he has given his estimate based on his experience of the tool he is using.
If you owned a torque wrench that you knew under read by 3lbs/ft what would you say if someone on an internet forum said ‘that's wrong mate – my different brand of torque wrench reads under by 10lbs/ft so that’s the true reading, yours must also under read by 10lbs/ft’ – hopefully you would tell him to p*** off!
In summary – comparing the figures from one chassis dynamometer to another is completely fruitless. Comparing a chassis dynamometer to an engine dynamometer is also completely fruitless.
The flywheel corrected power from a chassis dynamometer is obtained by using ‘fudge factors’ – as long as they are consistent on a particular chassis Dyno then they give really useful comparative numbers for that Dyno but that Dyno alone.
Even if using SAE or other correction factors then these are only valid within a certain range (SAE corrects for 15 deg.C to 31 Deg C). Anything outside of this range the HP figures can not be compared (So if your car was tested on a day where the ambient room tempreature in the dyno cell was below this the corrected figures will still not be accurate.)
To finish’ here is a quote from an article on drivetrain losses and HP readings:
In the end, there's no easy way to estimate the drivetrain loss your vehicle experiences on the road or even on the dyno. Coast-down tests are sometimes used on a dyno to attempt to measure frictional losses, but because this test is not dynamic (meaning they're not done while accelerating, but rather while coasting to a stop with the direct drive gear engaged but the clutch depressed so that the engine and transmission aren't linked) it really only captures steady-state drivetrain losses as well as rolling resistance. So rather than attempting to convert your vehicle's dyno-measured wheel horsepower to a SAE net horsepower figure using a percentage or a fixed horsepower value, you're far better off accepting the fact that these two types of horsepower measurements aren't easily correlated and forego any attempt at doing so
Full article here:
http://www.modified.com/tech/modp-1005-drivetrain-power-loss/viewall.html
No matter how much I write on this subject I know within weeks that someone will post 'my car made xxx HP at the wheels/flywheel today' and people will compare it to what theirs made on a different brand of dyno on a different day at a different location - please, please, please - it is a complete waste of time trying to do this!