XxOLIVERxX

Hello everyone
This thread will be devoted to further developing the H8 based ecu.
Hitachi Renesas H8" MH7202F & MHF203FA CPUs.

Found in the following models: 1996-2002
-Lancer Evo 5 6
-Lancer/Mirage 4G93
-Pajero 4G93 and many more

25570100 = Ceddymods MAF version based on ralliart rom.
25570300 = Ceddymods SD

I have been testing both roms and have successfully setup the current tables to "convert" the evo rom to run on my test subject mirage 4g93 with h8 ecu.
fuel trims never stray from 4-5%, the performance compared to the standard 4g93 rom, the new maf rom is almost the same, with the added features of the ceddymod rom.
im very pleased to say it drives well considering that i have not accounted for the fact that there may be major code differences between a mirage and an evo.

however, there are a number of things iv discovered about the behaviour of these roms in my quest to "add features to an otherwise stock car"
me being pretty picky about drivability i wanted to ensure the setup can be tuned correctly before adding a turbo kit.
i feel that there are more corrections to control the fuel and timing that have not yet been defined and i wish to discuss with others with the knowledge to find them and finish this rom off once and for all!

My only concerns with MAF rom are:
-EVO6: boost control seems to be working but having issues with boost cut intermittent -10 timing and fuel cut even though all the fuel cut settings are turned off/raised to the max. not sure if iv set something wrong or there is an issue with the rom? i found the other BDEL and TWGDC tables after digging through xmls, added those and chnaged those to match, still having the issue. not sure why these were not previously defined.. missed maybe? disabled/null? not sure.

-Mirage 4G93: the low load driveability is sometimes not quite right, when getting on and off the throttle it isn't as smooth as the original rom..
despite this it still drives well enough to not complain too much, but i'm interested in finding out if there are tables that are used to adjust the fuel finer than what we currently have.

So i copied all my settings over to the SD rom. with a few exceptions. (maf size) changed scaling and scaling adder/smoothing until it was close at mid/high rpm wot. i then built my ve table by logging and adjusting to get where i currently am.
however im now kinda stuck on a few issues. -cruising and idle is a massive headache!

Concerns with SD rom:
-Idle control issues, randomly bounces between 950 and 1150. doesn't seem to care what you set in the idle tables. even in closed loop it will not drop below 46 steps and does not change with the jagged rpm so we can assume it is not the cause. i assume its something to do with closed loop or idle learning. cutting spark or timing to change rpm.

ISC steps are solid at 46 and AFR is wandering with no evidence of timing change, load or ve change. ipw will change slightly by a tiny amount sometimes when it coughs/resets idle or whatever is happening.
when idle stabilises it can achieve target idle afr for 4-20 seconds and then either swings rich, or lean, sometimes both rapidly going from 12.5 to 16.5 when it calms down it hits the 14.7 i have tuned the ve table for and all is well. then the idle randomly goes up or drops and then it starts to swing the afr again, once it gets bad enough it will effect the kpa and will cause a change in ve which makes it worse.

After tuning the maf scaling adder in high hz i have target afr pretty much on point. above35% throttle it runs great. doesnt seem to do any of this afr swing stuff. until you start cruising at low load it will do the same as idle except while under partial throttle (15-30% approx) rich, target, lean, target, rich. over and over again.

So basically, if your never driving it slow ever, works fantastic. but trying to drive smoothly in a carpark or heavy traffic sucks.
long continuous straight roads you will feel a very slight jitter and the afr will swing occasionally, but adding 1% extra throttle usually kicks back to where it should be again.

off throttle acts as you would expect. can be tuned by ve table and it seems to follow it accurately on decel/zero tps. i have set it to not fuel at all and after a very long decel it did not kick on the injectors or anything odd that the normal settings wouldn't do.

so this brings me to the conclusion that the issue is defined to the following conditions.
-no throttle idle or part throttle cruise
-between load idle - 70ish. anything higher seems fine.

im wondering if the implementation of the SD has somehow "damaged" the routine in ways that will always cause this issue? and also maybe the idle control?

i can post logs, XMLs and rom files to anyone that wants to test or help look into whats happening.

i have a few cars i can use for testing.
stock ish evo 6 with 9turbo. will convert to SD soon
2 identical mirages with 4g93 engine. one setup for sd, one for maf.

iv researched endlessly trying to workout the cause and solution to these issues but im hoping to gather all the relevant info in one place!
if we can pull all the info together, port a few things over from other roms i think the H8 can be an awesome platform.

thanks for reading and i hope to hear back from some of you, hopefully we can build a little community since there isnt many people working with H8s at the current time

MinusPrevious

iTrader: (3)

Looks like a well laid out thread @XxOLIVERxX

Thanks for putting this together

Cheers

rjwkerkhof

Already asked you for the complete logfile of your boostcut event with the E6