CSL '0401' Program Binary Disassembly Notes

[karter16]

I'm so stoked too! Looks like it's outputting:
Intake Air Temp and Radiator temp: subtract 48, just like DS2 message
EGT: basically leave as-is, unlike DS2 message (multiply DS2 by 16)
Relative Opening: not sure yet, some scalar that looks the same as DS2. DS2 multiply by .0030518
Lambda Integrator: Subtract 2^15 then multiply by some factor that I'm not sure yet

https://datazap.me/u/bry5on/karter-can-logs?log=0&data=13-14-15-25-26-34&solo=15-17&mark=905

Awesome! I've got the config from TestO with the conversion factors so will dig those out shortly


Sent from my iPhone using Tapatalk

[karter16]

Okay - so all are served as per DS2 with the exception of TABG as you note.

TAN (Intake Air Temp): correct - subtract 48
TKA (Radiator Outlet Temp): correct - subtract 48
TABG (Exhaust Gas Temp): correct - leave as is, DS2 divides by 16 to fit this into a byte instead of a word.
AQ_REL (Relative Opening): multiply by 0.003051757
LA_F_REGLER1 (Lambda Integrator 1): multiply by 0.000030517

Let me know if this works :-)

[Bry5on]

Okay - so all are served as per DS2 with the exception of TABG as you note.

TAN (Intake Air Temp): correct - subtract 48
TKA (Radiator Outlet Temp): correct - subtract 48
TABG (Exhaust Gas Temp): correct - leave as is, DS2 divides by 16 to fit this into a byte instead of a word.
AQ_REL (Relative Opening): multiply by 0.003051757
LA_F_REGLER1 (Lambda Integrator 1): multiply by 0.000030517

Let me know if this works :-)

Beautiful. Thank you again my friend. Will test lambda integrator in the morning as it looks like I got the others sorted .. well, close on relative opening

[karter16]

awesome! let me know tomorrow if there's anything else I can do to help figure those out :-)

[karter16]

well, close on relative opening

Oh sorry! I completely missed that you already had that one - I wasn't meaning to be pedantic about the 7th decimal place lol

[Bry5on]

Oh sorry! I completely missed that you already had that one - I wasn't meaning to be pedantic about the 7th decimal place lol

Hahaha no stress. And I’m going to be taking this to 7 decimal places now. Because why not

[terra]

Looks like a lot of work and knowledge developed here since I became less active. Very cool! I guess I know what my reading will be tonight

[karter16]

Okay here's one for those who are concerned about heat-soak and the location of the IAT sensor.

The CSL software has additional code in the function that calculates TAN (intake air temperature):



Lines 16 to 26 are not found in the standard euro software. These lines are calculating a version of TAN specifically for the purpose of being used in the CSL's calculation of air mass, which in turn is used in the calculation of final RF as we've seen previously.

k_tan_m_cfg is set to 1 in the 0401 partial:



which means that tan_m is set from what I've tentatively named tan_m_adj.

How is tan_m_adj calculated? Like this:





The key thing to understand about this function is that it is iterative. It runs every 100ms and the previous result is used to calculate the next.

So what are the key behaviors?

To simplify it down a lot, the function basically:

1: Starts with IAT (measured intake temp)
2: Compares it with TMOT (motor temp) and blends it to some degree based on:
3: ML - load

At high load tan_m_adj is pretty just TAN, at low load conditions where the engine is idling the tan_m_adj function is distrustful of what the IAT is telling it and calculates tan_m_adj more and more heavily based on tmot the longer the conditions continue.

Why is this model needed? It is specifically to deal with heat-soak at idle and at stop conditions. The model prevents situations where the DME trusts the IAT in heat-soak scenarios. This would be bad as it would result in lean running conditions. The tan_m_adj model gives the DME a more realistic understanding of the actual intake temperature to ensure that lean conditions don't occur.

It's a bit hard to graph, but this kind of shows the response of the model at various levels of load:



This isn't completely representative of how it works in reality as it is based on a fixed TAN reading, when in reality TAN will rise quickly due to heat soak, but what this graph does show is that at higher load levels the model heavily favors actual TAN (higher airflow = more accurate reading) at lower load levels it skews towards TMOT over time.

I can't come up with a good way to graph or visually represent an actual scenario of a car coming to stop at the lights and the behavior of TAN and tan_m_adj, so you have to just work it through in your mind.

Essentially this a fairly complex model which is tuned to handle IAT heat soak scenarios and give the air mass calculations a more accurate IAT value in those scenarios. The model is iterative and adjusts the tan_m_adj value further towards TMOT the longer the heat soak conditions continue.

What's the lesson? Don't mess with your IAT sensor - the software is specifically calibrated to deal with heat soak conditions.

[heinzboehmer]

Custom DME FW is working great! Just ran through a session of VE tuning on my car and it was so nice to be able to pull all that info over CAN.

Feel free to peruse: http://datazap.me/u/heinzboehmer/csl...?log=0&data=15

Everything in that log is from CAN, except for throttle pedal position and the two brake pressures.


...which reminds me, karter16 any interest in adding a second CAN message with throttle pedal position? Having a reference of what you're telling the car to do with your foot is really useful. The ITBs and idle air valve on this engine (and some other stuff like DSC) make it hard to interpret the TPS logs. You never really know for certain if the throttles opened as much as they did because that's what you wanted or because the DME decided that that was how much they were gonna open.

[karter16]

...which reminds me, karter16 any interest in adding a second CAN message with throttle pedal position? Having a reference of what you're telling the car to do with your foot is really useful. The ITBs and idle air valve on this engine (and some other stuff like DSC) make it hard to interpret the TPS logs. You never really know for certain if the throttles opened as much as they did because that's what you wanted or because the DME decided that that was how much they were gonna open.

Yes! I've actually already been planning to look at adding a second message. I think it's going to be a bit harder as, unless we can conclusively prove one of the 14 other messages is actually not used then there isn't a spot for it, which would mean additional work. I have been doing more disassembly work this weekend around the wider behavior of the CAN system to understand what options there might be.

[Bry5on]

What are the other IDs? We can log them to see if data exists, although neither of us are SMG.

[karter16]

What are the other IDs? We can log them to see if data exists, although neither of us are SMG.

That would be great thank you! - I think the most likely ones are:

0x62F
0x700
0x701
0x770
0x513

[Bry5on]

Datalog shows only these:

[karter16]

Datalog shows only these:

Thanks very much for this. I've been doing some more investigation. I have more work to do to understand more but have figured out the below points:

• There is a second set of 15 CAN frames which are mapped to a separate region in memory
• There is a second set of read/write handlers
• These read-write handlers are more complex and appear to be built to manage the sending/receipt of a series of messages over CAN
• I'm not sure yet but it's possible that this is support for ISO 15765-2 (ISO-TP/CAN-TP) or similar for transmission of error codes and diagnostics over CAN
• This functionality is present for both the Master (main CAN) and Slave (SMG CAN)

I've also been trying to figure out CAN ID 0x770 as it seems to be paired in some way with CAN ID 0x771 (which is what I've repurposed for 0x7D0) I selected this one to repurpose as its handler pointer was to 0x00000000. But I've been thinking about that and I can't completely rule out that it's an intentional mechanism to allow reset of the processor via CAN. It seems a bit crude and would be a crash to a hard reset without any preparation, but maybe...


Anyway. In terms of the more immediate question of a second CAN message. I'm pretty comfortable that we can use the slot for 0x62F (read) as this message is only ever received for a chassis configuration that isn't the E46.

[heinzboehmer]

I selected this one to repurpose as its handler pointer was to 0x00000000. But I've been thinking about that and I can't completely rule out that it's an intentional mechanism to allow reset of the processor via CAN. It seems a bit crude and would be a crash to a hard reset without any preparation, but maybe...

Man, that would be a pretty crazy way to reset the DME!

Now I want to try and crash it over CAN