9 replies to this topic

[dosco]

Last year in a discussion on the Gas Savers forum, someone mentioned that fuel injection mapping in production cars was primarily controlled by intake air temperature. Part of the discussion was about changes in air density and impact on fuel efficiency ... for someone living at high altitude, theoretically the amount of fuel needed to achieve the proper ratio would be impacted (reduced) by the lower air density.

Any ideas on the specifics of how EFI systems control fuel mapping in production cars?

[NeilR]

I would not be surprised if the major manufacturers did not know the VE of their engines to less than 1% from 0-5000m and under the majority of atmospheric conditions - the maps used would adjust to suit the needs.

[gruntguru]

Last year in a discussion on the Gas Savers forum, someone mentioned that fuel injection mapping in production cars was primarily controlled by intake air temperature. Part of the discussion was about changes in air density and impact on fuel efficiency ... for someone living at high altitude, theoretically the amount of fuel needed to achieve the proper ratio would be impacted (reduced) by the lower air density.
Any ideas on the specifics of how EFI systems control fuel mapping in production cars?

Air mass flow is the principal measured variable - if the ECU knows the kg/sec of airflow, it can divide by the desired AFR and then knows how much fuel to add.
- Many systems do it directly with an air mass meter.
- Next best is an air volume meter (remember the Bosch vane units). This system will require measurement of air density (via temp and pressure sensors) to convert the volumetric flow rate to mass flow rate.
- Next best is a manifold pressure sensor (after the throttle valve) which will require a knowledge of engine speed, displacement and volumetric efficiency. The main shortcoming is that VE can change over the life of the engine (not insurmountable). This system will also need air temp and pressure sensors to convert volume to mass.
- Next best (worst of all) is the throttle position mapped system. Once again it assumes a VE, given RPM and throttle angle - fine until something changes - could be as simple as a dent in the exhaust pipe.

All the other sensors are of secondary importance and are used to trim the fuel under non-standard conditions eg cold engine, throttle transients etc. One exception is the Lambda sensor which has the final say on mixture under all operating conditions where emission control takes precedence.

[gruntguru]

I would not be surprised if the major manufacturers did not know the VE of their engines to less than 1% from 0-5000m and under the majority of atmospheric conditions - the maps used would adjust to suit the needs.

Volumetric airflow (and VE) changes very little with altitude. It is actually the density of the air which is different and this results in a change in massflow.

[J. Edlund]

Modern control units use actual airmass/combustion as the base for the fuel injection. This value is divided by 14.7 and then multiplied with several correction factors.

Airmass/combustion is also the base factor in the control of the electronic throttle, and the ECU has translation maps from airmass/c to torque which forms the basis of the torque demand structure. In boosted engines the boost pressure is also controlled by the airmass/c - it is adjusted so the airmass/c reach the desired value independent on the conditions.

Ignition is also controlled mainly by airmass/c and engine speed.

Actual airmass/c is measured by 1. airmassmeter and crankshaft position sensor (often with camshaft position sensor as backup for the latter), and 2. intake manifold pressure, temperature and the crank position sensor. The airmassmeter handles the general operation and the MAP based system handles transients where the airmassmeter due to the inertia of air will give incorrect readings. The MAP based system is also used as a backup in case the airmassmeter fails.

[Lee Nicolle]

On standard EFI road cars there is often bugs that are so hard to fix. Cars that go ok but drink fuel, cars that feel flat and doey. I have sent them to experts including major manufacturer dealerships that I know. Often they have no idea and want to give up.
I have heard and experienced putting resistors in temp senders to fool the computer, that is after replacing the sensor and numerous other dodges. I have had discussions with manufacturers technical reps and normally they have no idea either. Very frustrating.Often it is keep changing things until you get it right which can get expensive without a donor car to source parts from.
Programmable efi usually makes HP though often not driveability or fuel economy and emissions.
Personally I have driven similar cars in similar condition and got 200km difference from a tank of fuel, and the thirsty one had more power.
These cars often seem to be more outside temperature affected than some carby cars. And this on popular Aussie , japanese and korean cars. Everyday transport to most people.

[GreenMachine]

...Programmable efi usually makes HP though often not driveability or fuel economy and emissions. ....

Hmmm. Four variables there - what do you want, all four? Something is going to suffer if you do, you will have to set your priorities and then tune accordingly. You can maximise all, but not all at the same time.

Reminds me of when I was chasing some lightweight wheels for the racer, the saying was you can have cheap, you can have light, you can have strong - which two do you want?

[gruntguru]

Hmmm. Four variables there - what do you want, all four? Something is going to suffer if you do, you will have to set your priorities and then tune accordingly. You can maximise all, but not all at the same time.

You don't need all four at the same time. You only need power when the throttle is at or near 100% and the other three are simultaneously maximised at all other times (and power is only slightly compromised anyway).

Edited by gruntguru, 18 January 2010 - 11:10.

[Lee Nicolle]

I owned a Falcon that had been 'chipped up' with headers and big exhaust. It felt stronger than the normal cars but was very similar on fuel with normal driving except when you got into the throttle hard and then it got a bit hungry.
Though have seen other cars that are not very good at all except at full throttle.

[gruntguru]

it got a bit hungry.

Solid fuel?