Csl ecu conversion and software

[jylee05]

Understood, thank you heinz! 🙏

[heinzboehmer]

Thank you. So if I use Terra’s modified tune, there is no hardware I need to ECU related hardware I need to modify correct? I have a 2005 M3.

Yep, just make sure you follow terra's instructions

[jylee05]

Thank you. So if I use Terra’s modified tune, there is no hardware I need to ECU related hardware I need to modify correct? I have a 2005 M3.

[heinzboehmer]

Where can I find or download a stock CSL tune with the cam offsets changed for US 2005 Base M3 6MT?

You have to make your own. This thread has info about that: https://nam3forum.com/forums/forum/s...p-csl-h-bridge

Also, you can't just flash a CSL tune onto your DME. You need to BDM the entire CSL firmware on: https://nam3forum.com/forums/forum/s...ech-galletto-4

Or use terra's modified tune: https://nam3forum.com/forums/forum/s...asher-csl-tune

[jylee05]

Stock CSL tune runs perfectly fine in terms of driveability with just the cam offsets changed. However, your AFRs will be off in some places because the fuel maps are still setup for the CSL cams.

Where can I find or download a stock CSL tune with the cam offsets changed for US 2005 Base M3 6MT?

[e46m3cCSL]

TunerPro

Thanks! I will check with it.

[Slideways]

What program should I use for editing VANOS offsets like Cubieman did?


Sent from my iPhone using Tapatalk

TunerPro

[e46m3cCSL]

What program should I use for editing VANOS offsets like Cubieman did?


Sent from my iPhone using Tapatalk

[heinzboehmer]

Where are you guys getting this magical dissasembled euro SW version? Is this something you each created yourselves by running a binary through IDA Pro?

[terra]

The program you’re referencing isn’t 1801

These are the equivalent functions in 1801 with the correct labels - makes a lot more sense:

[SliM3]

Interesting, FF81C4 only pops-up @ loc_2599A & loc_2598E, in sub_2582C on my 1801 master. It makes sense because KF_LLS_TV adjusts idle AFR's, or least that's what I use.

1801 on left, CSL on right:






1801 on left, CSL on right:





Anyway, guess that explains why were seeing different things.

[terra]

Yeah that's generally my process too.

sub_1A406 in 1801_master is the non-CSL equivalent of sub_1a476 in the csl master. You'll see FF81C4 (rf_ti_const in 1801) is calculated the same way as FF81C6 is in the CSL software

sub_1a7a0 is the 1801 equivalent of sub_218d0

FFEDD2 is "tl" in the CSL software. Equivalent to FFEDEC in 1801_master

I don't know what FF807E is, I don't think it's "delta air mass" as you're calling it. Seems to be some sort of status byte that is populated with a value from a lookup table at 3D534 based on what's stored in FFEDD

rf_ti_const calculation
1801:



CSL:




Base Injection Time Calculation:

1801:


CSL:


I think the calculations are pretty clear

air density (K_RF_LUCHDICTHE), engine displacement (K_RF_HUBVOLUMEN), and K_HFM_TI_RATE are used to generate rf_ti_const

rf_ti_const is multiplied by rf to generate load / base fuel injector rate.

Edit: I see your mistake. You're cross referencing the 211323002001 binary, not 211323001801. The A2L you have is for 1801. RAM addresses can and do change between program variants.

[SliM3]

You have FF81C8 misidentified. That's rf_ti_const. Rest of it should make sense from there.

rf_ti_const is multiplied by rf in sub_218d0 to generate tl ("lastsignal" or "load signal" in english, which is the base injection duration -- units are ms). Mpower36 described how that is used in the slave above.

Take it for what's it worth but I use a disassembled Euro SW version and A2L to x-reference sub-routines and define RAM addresses. I don't have any reason to believe the CSL version deviates far from the Euro, since they have identical VE functions with the exception of a MAF

/begin COMPU_METHOD RF_ROH
""
RAT_FUNC
"%6.3"
"-"
COEFFS 0 1 -0 0 0 0.001
/end COMPU_METHOD
/begin MEASUREMENT RF_ROH
""
UWORD
RF_ROH
1 100
0 2
ECU_ADDRESS 0xFF81C6
/end MEASUREMENT

Cross referencing to the euro version, Sub_218d0 generates an ML value (Luftmasse or Air Mass) based on conversion from volumetric flow (RF) to mass air flow (kg/h), and is triggered by an engine condition that generated a DAM (delta air mass) value.

[terra]

Both my Master & Slave disassemblies are GTG! No doubt about that..




Again, anyone is more than welcomed to try it while monitoring the condition(s) where the AFR changes occurred. I can 100% guarantee you that it does not effect scaling of injectors nor base injector pulsewidth as I've already tried doing it that way in the past.

You have FF81C8 misidentified. That's rf_ti_const. Rest of it should make sense from there.

rf_ti_const is multiplied by rf in sub_218d0 to generate tl ("lastsignal" or "load signal" in english, which is the base injection duration -- units are ms). Mpower36 described how that is used in the slave above.

[SliM3]

A function of acceleration enrichment, triggered by delta air mass (DAM),