Motronic 2.3.2

[urquattro-island]

eeeh

say what???

[S4WILD]

по Русски и понятно

[UrS4boy]

по Русски и понятно

Which means (via bable fish): "in Russian it is understandable"

Duh.

[Jamo]

Appreciate the diversity of members here but can we try keep the posts in english not russian!

[Acidburn]

Appreciate the diversity of members here but can we try keep the posts in english not russian!

[Avtor]

So anybody can help?

[terrybullwon]

Unfortunately the information you are looking for is not free for distribution, the only people in reality, who could assist are Mihnea of MRC, S2central.Net, Wayne Schofield of Chipswizard & Mark of Automark.

As these make an earning from ECU programming your odds are slim. Good luck

Thats really the bottom line as Jamo says. These people make a living from this and are not going to give the info you need for free.

Or you could email bosch and ask?

[Avanter]

Try www.turbochip.lv may they will help you...

[porkkis]

Haha. I can't believe my eyes. Same story year after year. Here is my similar starting post 4 years ago: http://www.s2forum.com/forum/showthread.php?t=7075

I haven't asked Mihnea nor Paul nor Wayne nor Mark anything.

Somehow after these years I'm able to do good (IMO) 400+ tunes. Strange thing isn't it?

The map marked as lambda time 2 isn't used or if it is it's only in limp mode or something. The first one is. You can zero that and see what happens, very useful operation when fine tuning the lambda map after injector swap and sw scaling. If you don't do this the adaptation will screw things up eventually.

Basically the problem is that as these things are very hard and slow to learn and basically no one will help you, eventually when you know your stuff, you won't be telling it neighter.

Start by figuring out the few bytes before every map. You also might want to download the demo version of EVC's Winols SW.
A tracing emulator is a must for learning if you don't go the route of disassembling the whole crap.

Woops that basically was a bit of free information. I'm extremely sorry. You're welcome to blame me when M&co are going bankrupt because of me! The big money in chiptuning business is somewhere totally elsewhere than in these museum ecus and few jobs a year they give.

[S2central.net]

LOL @ 'museum ECU' - quite true, but the learning curve on this stuff is rather steep... That's as much the reason for those that learnt it the hard way not wishing to give anything away - as you said yourself. The MAF scaling thing is what I want to crack next, but I know its gonna take a while.

[porkkis]

I can imagine that you, aswell as I, are able to linearise the maf in about any other bosch ecu than M2-series.
I recently tried a spq maf on my own car. I got it working "somehow" without any special tricks but honestly it can't be done properly unless the linearisation can be altered. The basic ouput level is nearly 1V higher and different output curve causes great problems - also with resistor mods.
If you get it working I'm willing to BUY the information. I should probably find the motivation somewhere to start dissassembling the load routine once again.

Learning curve is steeper without D-files. I should probably say D-papers in this case. I really really really would like to have 'em

I think I already gave the advice how to start.

[S2central.net]

D-papers... yeah we wish

Maybe we can crack this MAF thing together then - I'd like to go with Bosch HFM2 or HFM5 rather than SQ hotwire - the HFMs at least have a documented profile. I had also assumed the shape of the curve was programmed into the ECU somehow... Lets take this one up privately if you wish to pursue it. Trouble is thats its so damn time consuming slogging thru assembly code in the dark - and working alone its sometimes easy to miss things.

[Aurélien]

Use a reverse enginering software .
Convert the scalling to floating point instruction , why that way you must find what map is it ...

@S2central , what's your problem with the maf ?

[S2central.net]

The problem is that S2/RS2 MAF has a curve of certain shape for mass of air vs output voltage but high power installs need a bigger MAF for improved dynamic range. These bigger MAFs have different shaped output curve. I haven't yet found how the shape of that curve is described in the maps so that a sampled voltage from the MAF correlates to the correct amount of air.

I know how to change the scales on the maps with MAF voltage, but that alone isn't very helpful.

[Aurélien]

I assume it's as on subaru's map ...

Your problem is for rescalling , or flow limitation in the ecu ?