How not to fake fuel consumption "measured" with ECU or onboard computer

[ausias]

I started a thread of How not to fake ECU fuel consumtption numbers, because doing some tests i found that there are little changes that produces large variaations in ECU OBDII fuel consumtion figures (or MFA or Scangauge)

Lets do a collective Research and assessment:
Real or fake fuel consumtion changes?

How not to fake fuel consumption "measured" with ECU or onboard computer - TDIClub Forums

I think there are some things that fool the estimated numbers related to fuel economy in the ECU:
- Hammer mod or the VCDS Injection Quantity idle adaptation.
- Correction factor present at least in SDI and VP engines tuneable with vag com.
- flashed ECU when values are not between non tuned factory limits.
- Crank Angle / Torsion Value mechanical adjustment PD.
- injector nozzle substitution without ECU remapping (even remapped requested and real values out of specification could be wrong)
-...


What do you think? are there more ways to fool the figures?

do you think a links compilation of each one related discussion is needed?

do you have precise fuel consumption measuring to compare before and after the mod?

does anyone have data logs (vcds, vag-com,...) and real fuel consumption numbers?

all my data logs and onboard values changed when I changed any of the above things, but I could not appreciatte any big differences in fuel consumption at fuel station. Full tank Liters divided by kilometers did not reflect fuel savings from 7.0 L/100km (33.6 us mpg) to 5.5 L/100km (42.8 us mpg) as computer said with angle synchro, i. ex.

There are lots of people who:
"Believe in chiptuning" Some say a good chiptuning saves fuel. Not mine, and I payed a ECO ecu tuning wich advances timing and will save 0.5 to 1.0 L/100km.
"Believe in magnets" as fuel savers of -20% (+25% mpg)
And there are a lot of Believers of savings witouth scientifical assessments of the causes that allowed the new fuel economy.
"Believe in premum fuels" "Believe in xxx gas saver"

I'm interest in experience (and detailed description), data, engine theory, engeenering sources, autodidactic research, etc. I will contribute with my own tests and data if needed.

What people believe in or thinks about this issues is not welcome but explained hypothesis. We need to focus.

The ECU does not measure directly injected fuel quantity nor the CO2 present in the exhaust so there is a good calculation that evaluates an aproximation and almost real fuel consumption when sensors, mechanical adjustements are in a exact point, etc. I tried to do an BSFC map from data logging and I got some big surprises with the numbers compared to what books say.

If we could found the exact way OBD (or vag) values like: IQ mg/R, torque Nm, L/h, calculation is made will be perfect.