What are the two curves? AU, EL, EBGT etc, Cobra? What are the Y axis units & conversion factor back to kg/hr? Do you have the formulae for each curve (Volts v's kg/hr)?

The ecu actually calculates the fuel (pw) side of the AFR. The air mass is a series of constants defined by the MAF transfer function table (& interpolation between data points) based on the sensor voltage output, with slight modification applied for varying density v's temp. These "constants" can obviously be intentionally skewed by the programmer which is what I think you are saying?

No that's not what i'm saying.
The two curves are the factory 55mm and 70mm used in EL’s. I have the Y-axis Kg/hr figures but, for this discussion you don’t need to know them. All you need to know is that they are the same.

It’s the way the different EEC’s interpret the programmed transfer function that’s important. Instaneous Mass Air Flow is integrated by the different EEC’s in different ways. This is why you can’t just take a transfer function from say an EB EECIV and put these figures straight into an EL EECV.
If you look at the the transfer functions below you’ll see what I mean. The Blue is the EL GT 70mm, the Red is the stock EL 55mm, and the Black is the 55mm from an EB.

You are a knowledgable and experienced guy and have helped many members of this and other forums, myself included. But on this one I just can't accept what you are saying as it is contradicated by real world situations.

The key facts are that the same MAF housing & sensor is used for EB1 - EL2, the engine specifications are virtually identical, the power output is the same and the emissions are similar. So the overall ecu calculations performed on the respective MAF transfer data tables (which are also virtually identical, excluding the special editions) to ultimately calculate injector pw & timing (through load) MUST ALSO BE virtually identical overall. This is an inescapable fact, and invalidates your second graph which shows dramatic differences in housing & sensor responce to airflow.

Now here are the MAF transfer functions from both (EB top, EL bottom). Note the deviations are much much smaller than your data shows. It is a slight tweak of around 5% leaner for the EF-EL rather than the 100% difference (ie double) that your data indicates

0.5 1 1.5 2 2.5 3 3.5 4 4.5
11.64 31.77 63.46 110.27 175.77 263.52 377.08 520.02 695.90
11.04 30.08 60.08 104.41 166.42 249.50 357.00 492.30 658.76

If your data was correct the use of the same housing & sensor on EB v's EL would roughly cause fuel to be used at double the rate. Or explained differently, it would definately preclude the use of the same housing & sensor on both ecu's. This is not the case, as noted above the same housing & sensor is used for all these models.

Further, to suggest a stock 55mm diameter EL MAF housing & sensor has a transfer function that goes all the way up to 1400kg/hr make no sense. It would imply the sensor only has a working range of approx 0.5-3.75V because it sure won't flow anywhere near 1400kg/hr! An approximate airflow calculation based upon a 55mm housing flow potential of 275HP, BSFC of 0.5lb/hp-hr, 12.5:1 AFR, 2.2lb/kg gives a mass air flow maximum around 780kg/hr. Even a 73mm housing (EBGT/Sprint/ELGT/AU) is limited to around 1000kg/hr. To reach a flow of 1500kg/hr as your graph shows would require a housing in the 90mm range.

Integration is simply calculation of area under a curve. There is only a single answer to this for the same curve. Different methods can be used to reach the answer, but it will always be the same.

Now I'm not sure what the relevance of the first graph actually was? You say it compares a stock EL 55mm housing & sensor with an ELGT 70mm housing & sensor, using some underfined units. But this thread had morphed into comparing a stock E-series 55mm housing & sensor with an AU 70mm housing & sensor / E-series sensor.

What I can say is that in my experience trying to run an AU 70mm housing & E-series sensor on an E-series ecu has been problematic, leading to detonation under load which required either a reduction in base timing or higher octane fuel to suppress it. Swap the stock unit back in, and the problem goes away. Clearly a lean condition, which as we all know is a dangerous operating condition for a petrol engine.

I have knocked up a MAF plug adapter and I will try an AU 70mm housing & sensor in my car next time I get one. I'll also swap out the sensor for an E-series and try that as well. I have a wideband installed and it will answer this once and for all by removing all the variables.

I don’t know how many times I have to say, ignore the numbers!…it’s just a representation of the differences between the MAF transfers because it's the only way I could get a print screen to work.
But ok, let’s use the figures you have listed. …They are still different are they not? Now why would an identical MAF have a different transfer function just because it’s in a different EEC?.... Answer; Because they calculate air flow differently! Yes Integration is the calculation of area under the curve for the average air charge and, this calculation is done differently between EEC’s! There will only be a single answer for the same curve in the same EEC but the answer will be different if the same curve is put into a different EEC! It doesn’t mean the end calibration of the engine is different. It does mean however that you cannot simply transfer a MAF transfer function from one kind of EEC to another.

And listen mate, by all means go off and try this AU MAF on an EL for yourself but don’t ever suggest my experience is not “Real World”