Page 1 of 1
interpolation, TP values, constants etc
Posted: Tue Jan 13, 2009 9:44 pm
by chris2712au
hi all,
Trying to get my head around the bits and peices..
Heard alot about resolution of different ecu's and maps etc..
however with the nissan system it is calculating load based on afm voltage, rpm, and injector pulse ?
then as for the maps is it averaging all four cells in grey ?
I have been just going in and changing it to what I need but would like to know a bit more about the nuts and bolts of the system..
Posted: Thu Jan 15, 2009 1:02 am
by Legionnaire
ditto that. I would also like to know how the value from my fuel map is translated to actual injector pulsation? particularly in ECUs with "target AFR" and volumetric efficiency tables.
Posted: Thu Jan 15, 2009 9:38 am
by kenbam
I'm pretty sure the target A/F ratios on the maps are only a basic estimate based on raw values, likely never get the actual A/f ratios readings to match the map . the 4 cells highlighted are interpolated, so the actual injection IS an average of all 4 cells.
Injector pulsewidth will also never match whats on the maps as they are a base setting. Adjustments are made (trims) to the base settings based on things like engine temp, ambient air temp, engine load and long and short term A/F fuel learning adjustments (adjustments made to fuel based on O2 sensor).
Posted: Thu Jan 15, 2009 10:28 am
by MichaS14a
That is the reason why I asked if someone can disassemble the programcode from the ECU. I this will happen there is no more miracle in what the ECU is doing and why.
I am also interested in the formulars to calculate the right injection pulse with all correction parameters.
Posted: Tue Jan 20, 2009 12:07 am
by Matt
people have disassembled the ECU program code in the Nissan tuning community. I've done bits and pieces where interested in sussing out how things work in more detail
the four grey cells are 'read' and the interpreted value based on RPM and TP values is calculated as a result. the closest value used is the darker grey value cell out of the four
timing map is a good way to see this along with the consult timing guage. you will notice the interpolation working as the cursor moves through the timing values in the map against guage values
the target AFRs are purely theoretical and are subject to change after MAF and injector resizing substantially. they are good to get an idea, but a lot of people use filtered values and monitor the output AFRs
i've got an open action to allow fudging of the AFRs to make it represent actual read (wideband) ones but not implemented yet. Using the injection pulsewidth guage is a good way to see changes to fuel maps working as you expect
Posted: Tue Jan 20, 2009 4:45 am
by UNISA JECS
Its pretty easy to get your actual wideband AFR's matching the theoritcal AFR's within reason. Injection Multiplier is yoru best friend here for WOT adjustment and for closed loop Injector Latency and and TTP Min. are you best friends for the idle and closed loop low throttle contions.
P.S. I have my wideband AFR's matching my theoritical AFR's within reason .5 AFR at the majority of situation, there still needs to be some fine tunning done.
Posted: Wed Jan 21, 2009 11:24 am
by Legionnaire
What about VE tables then? How is target AFR converted into actual AFR? Is the VE table value a multiplier of some sort, e.g. 104% works like target AFR*1.04?
If I am sorting my fuelling from scratch, what is the proper approach here - to set all VE values to 100% and then tune target AFRs or to set up AFR table where I want it to be and then fine tune with VE map? Or should I combine both methods using #1 for raw fuelling then #2 for fine tuning?
Posted: Wed Jan 21, 2009 1:19 pm
by Matt
More documentation required... and I'm working on it. In the interim we will cover things here and I'll use it as a basis for the documentation
Basically the VE table interpolates the same as fuel/timing maps. It will take the average of the four cells (based closest cells to fuel/TP values determining row/column) and then use that as a multplier for the injection pulsewidth in conjunction with the values in the fuel map
So change either table and your injection pulsewidth will be modified
Now as for tuning them from scratch. Set your fuel table to the desired AFRs that you are tuning towards, and then modify your VE tables until your wideband reaches those AFRs.
You may wish to set them to 100% to start with across the board and then modify as you go. I'm working on autotune for VE but as the scalers are different to the fuel map (smaller area) its still requiring a bit more work to finish that