ecu fuel trim

[jondee86]

It seems strange that the ECU does not have a target AFR table hidden
away somewhere ?? I was under the impression (from reading about the
Freedom ECU on club4AG) that it was capable of self tuning a number of
selectable maps according to driving style e.g. economy, normal and
sports modes. To me this implies having a target AFR table based on
TPS x RPM.

I gathered that once these maps were auto-tuned in closed loop, they
could be locked in, and later used in open loop. Maybe I read too much
into that article ? Or maybe you have a different version ?

My ECU (which is not super-sophisticated) can only store two maps. But
it will auto-tune to redline in closed loop, using the target AFR table. There
is an option to go to open loop above a specific TPS (maybe RPM) and use
whatever values you have setup in the map.

Seems to me you might need to dyno tune your WOT values using a
"safe" AFR. Then manually taper the map back intuitively to your last line
of auto-tuned cells. Fine tune by monitoring from your logs while running
open loop, and then lock it up

Cheers... jondee86

[totta Crolla]

Hey Jondee yeah you are right about some versions of the Freedom
having target AFR values, unfortunately though mine does not have this feature
I guess the sport mode is the "throttle map" mode that uses TPS x RPM
and the economy mode uses the MAP sensor.


"Seems to me you might need to dyno tune your WOT values using a
"safe" AFR. Then manually taper the map back intuitively to your last line
of auto-tuned cells. Fine tune by monitoring from your logs while running
open loop, and then lock it up"

I'm doing this at the moment, the ecu allows a percentage of fuel to be added or subtracted from the whole map so now I have richened the whole map by 5% allowed learning (14.7) for 60% of the map but I'm
still safe at WOT and full load by virtue of the 5% enrichment.

Sending the ecu back to the manufacturer for the full learning modification
is also an option.

[jondee86]

By coincidence, last night I read an interesting comment regarding the
Freedom ECU. It seems that when in self-learning mode, instead of the
ECU waiting for the load/rpm combination to place the cross hairs near
enough to the centre of a cell before making a correction, it takes a value
at random, and interpolates proportionately between the four nearest cells.

Adjusting cell by cell takes a lot longer, as the cross hairs have to linger
near the centre of the cell for a few milliseconds, and this does not happen
so often. Interpolating between the four nearest cells can always be done
with no waiting time.

I didn't explain that very well, but it ties back to your original post

Cheers... jondee86