Modifying Motronic 2.3.2 ECU hardware and software.

[-ice-]

Hi,

The map i'm looking at is 0x8cb8 (551aa), 0x8ccc(rs2 551b) or 0xacdo(rs2 551c)

0x2e is fed with the accumulator, which is loaded with the xram address 0x6b before. And xram 0x6b is filled from another location also with RAM_2e. Have to trace it back...

Might have something to do with load, because its compared to 3f after the map access.

But i don't know yet.


Regards

[kymi]

The axis shuld be 0x16F (xram) and it serves a purpose (look in the diag-routines).

[-ice-]

Correction: The address 0x8cb8 for the 551aa file is wrong. It seems 551aa doesn't have this 5x4 Map.

Bye

[kymi]

rs2 551c, 0xACD0 should be at least pretty straight forward with axles and what it does.

[-ice-]

Next correction

What has been 5x1 in 551aa seems to be 5x4 in RS2 551b & c.

In the 551aa file the last bits of RAM_2e are used to switch the maps. In this case Map 0x8cb8 and Map 0x8cc4.

I guess this two maps are load limiters or something like that. Second map perhaps for an emergency condition.

So RAM_2E in this case is used bitwise to select something, it's not a variable like rpm, load etc. That's what I think at the moment.

[kymi]

rs2 551c feeds the temporary 2e with xram 0x16F for this map read. You're on the spot.

[kymi]

Ok, I was a few steps after your spot.

In rs2 551c xram 0x16B.1 bit is used to check should we run 0xACD0 (5x4) or 0xACF1 (5x1) map. Look at the maps and routines run after the map lookup, it'll explain.

[-ice-]

The carry bit is used to compare it with load. Next map used is 0xACFD (Also with this Ram 2e Axis)

The procedure is called in this background routine, where also the MAF linearization is called.

What do you mean?

Bye

[kymi]

I mean that after the map access the result is cjne'd to ram_3f in both of those cases I mentioned. You have two situations doing about the same thing, only the data and axis (and 2d/3d) is different. Why the other table is using only rpm and has the values it has, and other table is 3d with different values?

0xACFD has the same ram_2e axis xram 0x16F, fed just before the map access.

If you have access to tuned binaries, check out the map ACD0.

[-ice-]

The 5x4 Map ist the one which needs to be tuned. In the stock bin these 4 lines (displayed in 3D) are flat.
Remarkable is also that in the 551aa file the 5x4 Map does not exist and is only 5x1. So the RS2 bin can do something, which the old 551aa ecus are not able to do.


But what it actually does... Results are compared with load(3f) so it will limit something.

[kymi]

Just look closer and check the diag for 0x16F.

[prj]

Come on, it's not that difficult.

551B:

5x4 is the casual load cut, RPM x Baro pressure.
If load is above this for 0x8389 time, fuel is cut and boost disabled.

After that the 5x4 limiter is ignored and the second limiter is used instead.
That limiter does not have a delay. If load goes above it, fuel is instantly cut. This should never happen normally, because the first limiter disabled the N75. So if this limiter is being hit, it means we have serious issues with boost control going on, thus fuel is instantly cut to prevent damage to the engine.

There, now move on to more productive things

[-ice-]

prj, you should write a book about motronic 2.3.2

I could confirm the barometric sensor with an obd group (same xram address is recalled there)

Next point concerning the fueling (map read) procedure.

The first map read there is 4x1 with rpm axis. It's usually flat for a wide rpm range. I thing this might have to do with fueling when throttle is closed. Or am i completely wrong?

Regards

[-ice-]

I was not looking for it, but the Grundanpassungskonstante fgat0 just seems to be a hardcoded byte in the fueling-calculation procedures. Just a multiplier.

ByeBye

[prj]

I was not looking for it, but the Grundanpassungskonstante fgat0 just seems to be a hardcoded byte in the fueling-calculation procedures. Just a multiplier.

ByeBye

Touché.