There are two things we are looking at here;

1) Comparison of cars on different dyno's

2) Drivetrain loss calculation and what approximate flywheel calculation the S65 is making

So let's go over these briefly for now and any questions that follow we can then start looking at in some more depth.

**1) Comparison of the same car (E92 M3) on the same and different dyno**
I have found in the BMW community that not many dyno operators (and there are a few!) talk about this subject too much. I have spoken to dyno operators across the world and some have said they feel nervous about this subject because if people knew everything then people would not take them more seriously.

We think the opposite and have posted on many forums about the dyno. We believe people should understand. It is quite clear that this lack of understanding has created confusion in the community and worse still some people going to the point of saying dyno’s are useless.

I would hope that people would take our points seriously as we do dyno a very large number of M power cars all year round for quite some years now.

First thing first. We know that weather conditions have an effect on engine power. Humidity levels, barometric pressure and temperature all individually effect engine power.

The majority of dyno’s have SAE correction factors built into them along with a weather station which measures humidity, pressure and temperature. The values are input into a forumula and the end result is the correction factor.

The correction factor corrects for the atmospheric condition under which the engine is being tested under.

The problem we have found (as many other dyno operators have found) is that these correction factors assume all engines will respond in exactly the same way as each other given a change of any of the above mentioned variables in any permutation or combination.

This is not true unfortunately and this is part of the reason why we get varying numbers especially under extreme changes in weather condition where the correction factor will also change massively.

Here is an example of an S62B50 V8 engine (BMW E39 M5) tested on our dyno dynamics machine.

Exactly the same tyres and fuel:

Notice how the graph shapes are almost identical! Small differences are there because slightly different ignition targets are reached due to the ECU actually taking some of the atmospheric conditions into account but not as much as the dyno is assuming! The dyno SAE correction factor is simply ‘over compensating’ because what it thinks should happen simply is not happening with this engine. What the dyno software should be doing is showing the car to make the same power on these two very different days..... that's what correction factors are meant to do! This is not an isolated example. This is however an extreme example but you get the point.

We have also seen different brand of tyres, final drive ratio’s and of course strap tension cause skewing of the dyno graphs.

What also can make a big difference is the gear used and I know for a fact that some dyno operators prefer to use 4th and some prefer 5th. You simply cannot compare.

Fuel of course plays a major role but this we all know and don’t really need to discuss here as we are talking about the comparison of cars with the only variable being the atmospheric conditions (hence the correction factor) and dyno operator.

There are further variables at play too which I have discussed very recently which are the type of cooling fan used, allowing ignition targets to be met and generally allowing the inlet, oil and coolant temperatures to reach an equilibrium. We will discuss this in more detail.

Just from the above brief information I am sure you can deduce that the same car can read so so differently on even the same brand of dyno on different days! Even the very same dyno can have varying figures on the same car because someone decided to change from normal street tyres to semi slick. Yes gentlemen, this also makes a difference.

Luckily for many, we do have some excellent dyno operators around who are also lucky to have very consistent weather conditions.

Typically the effects of using different types of tyre, final drive and the correction factors playing a role will skew the dyno graph. The shape of the graph will not however change and more emphasis needs to be put on studying the shape.

What also needs more emphasis is the logging of ignition advance and AFR. Thanks to certain members of the community this has been highlighted but only a handful of people look at this information correctly.

The dyno should be used in a more detailed way and is really for comparisons of upgrades and more importantly from our point of view as a development and test rig.

**2) Drivetrain loss calculation and what approximate flywheel calculation the S65 is making**
There are many dyno brands which attempt to calculate the flywheel power of an engine from the rear wheel horsepower made.

Two such dyno’s are the Dyno Dynamics and MAHA.

Dyno Dynamics uses a simply percentage addition to the RWHP which is set by the shootout mode selected.

For the S65 it’s shoot 8 (meaning 8 Cyl NA engine).

The MAHA uses a completely different method where the RWHP along with the coast down frictional losses. These are are combined to work out the flywheel power.

Why am I talking about this?

Well, both read fairly close to each other in terms of RWHP and the calculated flywheel calculation. They also both read very close to the stated manufacturer stated power and torque if used correctly.

Let us take some examples from our Dyno Dynamics readings across different BMW models and from there you can make up your own mind if it’s realistic or not.

Average Dyno Dynamics readings on Evolve Dyno for different STOCK BMW models:

BMW E36 M3 3.2 (Euro)

Stated Power – 321PS / 316HP

Average Tested Wheel Result – 250HP

Flywheel Calculation – 310HP

BMW E46 M3 (Euro)

Stated Power – 343PS / 338HP

Average Tested Wheel Result – 265HP

Flywheel Calculation – 330HP

BMW E39 M5

Stated Power – 394HP

Average Tested Wheel Result – 325HP

Flywheel Calculation – 395HP

BMW E60 M5

Stated Power – 507HP

Average Tested Wheel Result – 408HP

Flywheel Calculation – 497HP

BMW E92 M3

Stated Power – 414HP

Average Tested Wheel Result – 335HP

Flywheel Calculation – 410HP

I can post numerous (hundreds for some cars!) dyno graphs of the above cars to confirm all of this.

What’s more interesting is that the dyno graphs related to the above tests show calculated torque figures to be very very close to the stated torque from factory which in my opinion really does suggest that the calculation of around 18-20% drivetrain loss is very very accurate. I have never ever seen anyone look at it from this angle and those of you who are intelligent will agree that if both the flywheel calculated HP and torque are close together then it’s a very good estimate.

Hopefully the above will generate some more questions and we can look at certain areas in more detail.