RomRaider

Found this in a mates Australian Delivered Bugeye WRX M/T today.

Any help appreciated EcuFlash DL/UL and Romraider Def's required if possible please

Cheers

Ticket opened: http://trac2.assembla.com/romraider/ticket/127

OK, so I had a go at defining this map myself. I was looking for the typical boost maps, fuel and ignition maps as atleast a starting point. I found them in a Hex editor but have been unable to get a 'normal' XML file from a known ROM to work. The XML I was using was the A4SE900I, with the hex value lengths corresponding correctly for X axis, Y axis and cell values, it still appeared to have the scales off a LOT.

Is there anything else I need to change in the XML to get it to work?

Are there similar tables names for the existing ROM but with a space at the end of the name? If so these typically refer to tables with different scaling values. To use them you adjust the name of your definition to have a space at the end of its name.
Sorry I can't help more but I've never looking into a 16 bit ECU ROM.

There is unfortunately, no similiar ROMs to the A4SE50TI so I am using a definition table from the A4SE900I as it has an identical match for the fuel and ignition table sizes and values etc, The Boost table was also easy to find in the hex editor. Yet when I input the new hex values, the defined file is corrupted in some way and the table ends up having the right half of the table on the left, then the left half on the right etc. Was weird... almost like I was cross eyed

So my process was opening the A4SE900I XML file in suitable viewer/editor
Open the A4SE900I in HxD and find where the cells in the XML refer to in the Hex file
Then look for these strings in the undefined file, which was no problem, easily found
Put the new hex vaules into a 'copied' version of the XML table and it does not work(problem as above)

Anyway, I am happy to define where the hex values are that refer to if anyone is able to get the XML working correctly

Can you post the corrupt xml to are trying to use?

I think I have worked this out, the ROM image is 160Kb but the 'donor' XML is for a 192Kb ROM image.

Working on solution now

EDIT: Done. It was the 160Kb to 192Kb conversion which was the failure point.

Some 16bit ROMS have a 8000Hex memory hole from 20000 to 27FFF that ECUFlash ignores when it downloads as it is RAM memory but ends in it being saved as a 160Kb file. I used HxD to insert the necessary zero values for 8000hex as a paste insert to all the values I had found to make it 192Kb. Result works

Attached is the EcuFlash Rom definition, I intend on copyng all the values into existing RomRaider definition file though

Nice work mate. I have an ADM 02 WRX with A4SE900I ROM ID, did you notice many differences between those two ROM images?

50TI runs less boost compared to the 900I

50TI runs the same ignition base timing but less timing from 2000 to 6000rpm in the timing advance tables above 1.0g/rev

50TI also runs nearly a full point richer under open loop above 4000rpm but runs leaner below 4000rpm... Hmm, actually comparing the 2 primary open loops and the 2 failsafe open loops, I am not totally convinced I have the 50TI Primary and Failsafe definitions in the correct order. The 50TI RPM scalings line up with the opposite version of the 9000I file. Surely Subaru would not run 8.75:1 AFR's on a Primary map and 9.85:1 on the Failsafe map though?? I am going back now to look at this further, I might have missed something.



EDIT: I just noticed that the 900I and the 50TI have the air flow meter Hex values as being identical(I somehow left the 900I MAF scaling table in place when I deleted the others) so I am going to experiment with restoring all of the 900I tables and see if any others of relevance line up

OK let me know when you've done that and if you could put up a new def that would be great

[e] Look at the X axis for those fuel maps, i think they're in the correct order.

Well, I actually found a few others that lined up on ROM addresses. Surprised at the number that did actually.

All the Mass Airflow / Engine Load maps lined up perfectly, along with Oxygen Sensor scaling and compensation tables and the IAT and ECT scaling tables.

Max Wastegate tables were only a few cells off. Pretty sure I can get the other 3D and 2D tables sorted

However, I flashed the car in question last night with just a revised fuel and timing map, which alone has made a huge difference.

PS: I have the Australian World Rally on right outside my house right now(in Coffs Harbour NSW) which is pretty inspiring haha

So would it be safe to flash a A4SE700i or A4SE900i ROM image over this image?

i flashed a4te100g (edm 03 wrx) over a4se700i - it's 99%
plus now i have launch control and "type 2" boost control (per gear wgdc, 3d initial, etc). had to disable a alternator cel (charges fine) and it runs on a bit after switching off. i'd say most comparable maps using wrx02 method will flash over without too much drama.

i actually (accidentally) flashed my a4se700i rom over an a4rm100g (adm 03 sti) and it still started and idled ok. it did chuck a cel too though, can't remember specifically - o2 sensor related

Thanks for the confirmation

I had a look at the stock ECU Base Rom listings, the 1B54500505 ecu ID is used on a few roms, the A4SE900i looks to be the most compatible and is also the rom image I used as a basis for finding the tables.

Failing that, the carberry project will have a new candidate haha