RomRaider

Quick question for you guys.

http://www.scoobypedia.co.uk/index.php/ ... yList#toc3 <-- I'm using the AF423 code defnition available here, and needed to shut off some codes on a 2.5RS/WRX swaped car. I know at least one of the codes is defined in the XML defs file, so that isn't an issue. But the other 3 I believe were not.

P0183 <--0x2D971 (tentative)
P0447 <--0x2DA71 (tentative)
P1230 <--0x2DB21 (tentative)
P0037 <-- the one supported for the AF423

Now, if I'm to take a guess here... from what I can tell you have an identifier of 03 or 04 qualifier, and then the code itself.

That being the case, if I were to fathom a guess, for P0183, is 0x2D971, or 04 0183. A search for 03 0183 proved no data, but there is one match for 04 0183. From what I can also tell, replacing 04 0183 with 05 0000 is how you disable the code.

Do I have the right idea here?

I want to make sure I can do this before purchasing a tactrix cable for this.

I'm not sure if the locations are correct but the disable is the same. What is the actual ROM ID? It may already be supported.

Well, without a tactrix cable, I can't do the dump itself, but if it's anything like using Flash2k2 and the Protuner, several different rom's can be burned onto a physical ECU (case in point, I used an AF423 in Flash2k2 on my AF411 ECU).

So, for sake of this instructional post, I'm using the AF423 I listed above as a test case. Also, I went through the AF423 definitions in the RomRaider distribution, and 0037 was the only one defined, of the 4 I need shut off.

af423 is A4SGC00C. This method does not work for all CELs. I know that it doesn't work with any of the TGV cels. Only way to find out is to try it. I've heard that the 04 and 03 refer to either generic obdii codes or subaru specific codes. The addresses look correct.

Interesting. Well, I'll definately order up a Tactrix cable and give it a try. Just to re-iterate, short of TGV codes, using the 04/03 qualifier method of identifying a MIL code is reasonable across Subaru ROM types?

Yes, but again, as far as whether it will work outside of the CELs already in the definition, it is hard to say without trying it. Another guy tried to get rid of a crank position sensor CEL and it didn't work either. You might also try 00 00 00, as this it what Cobb uses on some of the CELs. Not sure of the signficance, though, but they use all zeroes on p0139, p0137, p0038 instead of 05 00 00. I haven't set up the definition this way, as there have been no reported problems with the current method and you can see Subaru use 05 00 00 for CELs that don't apply to a revision.

This will only work with non-DBW ECUs, though. The CEL codes are always close to the end of the ROM on the 16-bit (I usually skip to 0x29000 to start a search). JDM roms reverse the first byte compared to USDM/EDM - for example, instead of 03 some of the same cels are 04. Not sure why, though and I haven't got into it enough to add cel defeats for the JDM roms.

It's likely that the 3 bytes represent a memory location and max and min values, somewhat like how the checksum fixes work. Whatever location 00 is pointing to would then always be 00 and never trigger a CEL.

Not sure what you mean, but the last two bytes are the actual code represented as an actual unconverted value in hex (forget what you call that). So, code P0420 is shown as 0x030420.

Thanks for chiming in Jared. See, even I have found a use for this technology.

Also, I did see your last edit on the announcement post on NASIOC. I do have a plethora of ideas, and I do need to find a moment to sit down and collate them all into text for you.

Good to see you here. I didn't really get the sense from your posts that you were anti-open source, and I can see where you're coming from.

Hey I think you're talking about my situation. It was actually the 'cam angle position sensor' that would freak out every once in a while. Strange thing as it does it very infrequently still. I can go weeks and not through one, but then yesterday it happened twice. I haven't been on here much in the past month as I've been tuning on the pre-release of TXS' new speed density firmware.

and sometimes it starts with 05 like the misfire codes i had. p0301,2,3,4 all started with 05 03 01, 05 03 02, 05 03 03, 05 03 04. i turned all of those into 05 00 00 and now i get a cel, but no code. i figure it just has no value so it tells the scanner theres nothing there, but whatever triggers the actual cel to come on is still working somewhere.