NefMoto

Hi

I've read about charge limitation based on gear. According to Gremlin's post, the KFMDBGRG (Map) and / or MDBGRGA (Characteristic line) can be used to achieve charge limitation based on gear.


Due to the influence of the variable mibgr_w on MDKOG, the use of MDBGRGA is preferred:



Can someone please tell me

• - at which address the characteristic line MDBGRGA in the 8D0 907 551 K starts?
• - size (I guess 8x1)?
• - factor (I guess multiplied by 0.001526)?
• - data organization (I guess 16bit)

There is no such map on the RS4.

If you want per gear on air path, you will have to modify the asm code for LDRXN lookup and move the map to the end of the cal area.

Hello prj

Thank you for taking the time to respond to my question and for being willing to share your knowledge. I appreciate that.  :)

I can not (yet) change or write assembler code for the C167. But if I understand the K-Box code of the ME7.1 very well, I would like to learn to write assembler code for the C167 with IDA Pro.

My goal is to reduce the exhaust gas temperature at gear 4-6 by reducing the target fill. But the solution must not leads to retarded ignition, because this would increase the exhaust gas temperature and I would miss my target...

Apart from your suggested solution of writing own code, there is no way to achieve my goal, right?

https://xyproblem.info/ (https://xyproblem.info/)

Use the EGT sensors on the car and set the limiter.
Or if you want a larger control area then fit the RS6 ones, they can read already from -40C.

The RS6 parts are:
077919529D
077919529E

There's aftermarket replacements that exist for them too.

Thank you for your advice. I have already read in the thread http://nefariousmotorsports.com/forum/index.php?topic=1998.15 about the use of RS6 EGTs.


OK, one step back ;)

Actual state (Problem)
If I made a full pull in 1th, 2th or 3th gear, the EGT is fine. But a full pull in 4th gear result in a little too high EGT (and logicaly even higher EGT in 5th and 6th gear than in 4th gear).


Prefered state (Goal)

• - Regardless of which gear is used for acceleration, the EGT must not become too high.
• - The requested load for the 1st, 2nd and 3rd gear should not be reduced.

Rejected solutions

• - KFMDBGRG (because the ignition angle is also affected)
• - MDBGRGA (because in the K-Box Flash is not available and I can't write own assembler code)
• - Run engine with even more excess fuel (lower lambda value) --> I don't want to go lower than the current lambda value
• - Increase ignition advance (not possible because engine knock)

What do you mean by "set the limiter"? I don't think you're talking about the rev limiter, are you?

This is the most optimal way of EGT control. You "don't want to go" - why exactly? When you are hitting high EGT then your nr1 priority is to limit it to save the engine. What you want or not want is not important at that point.
This is the exact reason why these cars have EGT sensors from factory, to not have to run very rich lambda in 1-2-3 gear, but to protect engine in case of very long WOT situation.

TABGSS of course.
All you need to do is set this value and then it will not exceed it by dumping fuel.

If you want to additionally to reduce the boost in this situation, then also configure LDOB (overboost function) to give you a limited time (in seconds) with the higher boost target, and then give a lockout after that.
For better protection with bad fuel there is additionally KFFLDEO, KFFLLDE, KFFSLDE. This is already somewhat configured from factory.

By using all these features together you can achieve very good EGT control, but nr1 is fitting EGT sensors if they are not already fitted, because this will save the engine.
If you configure LDOB and LDRLMX maps correctly then it will almost never hit the TABGSS limiter.

Any kind of gear based load limiters are the completely wrong approach. The EGT is not dependent on gear, it is dependent on how long you stay WOT and on fuel/ignition conditions. If you put the car on eddy current brake dyno, you can easily hit the high EGT in 3rd or 4th gear by making the dyno hold the roller speed constant.

I let the ECU enrich the fuel mixture with fuel, but the enrichment should never reach such low lambda values as this OEM is the case. The experience of my engine builder has shown that full load driving with too low lambda value can massively reduce the life of connecting rod bearings and the life of the oil (depending on how long the full load driving with very low lambda value is done and how much the oil is damaged).

Instead of going lower and lower with the lambda value and reaching OEM values, I would much rather reduce the engine load so that the EGT decreases. But unfortunately there is nothing out of the box for that. It only exist KFFKRXTM, based on coolant temperature but not on EGT.


You're right about that. But I think this can be neglected when driving on the road.



Thank you so much for taking the time to point out and naming all the relevant maps and their correlations. This is very usefull to me because it helps me check myself and my knowledge.


Comprehension question about TABGSS
If the EGT temperature exceeds the threshold of 979.99 degrees Celsius in TABGSS and the engine load is greater than the value in KFRLATR, the fuel mixture is continuously enriched, in extreme cases up to lambda 0.6? Are the limitations for lambda target (interaction of KFLBTS, KFFDLBTS and KFDLBTS) not considered?

Wrong. It will also go 0.75 or lower lambda OEM.
Your "builder" has 0 understanding of combustion process. On WOT the mixture richness is not going to do anything.
The problem is on cold start when the fuel does not evaporate. Nothing happens to the oil running rich on WOT because the engine is hot and unburned mixture goes to exhaust.  It does not get deposited on cylinder walls or in the oil.
Claiming this is ridiculous.

You invented an artificial "problem" and trying to solve this "problem".
Your solution is not inventing an artificial "problem".

There is a minimum lambda target. Overall you should tune the engine so that it does not need 0.7 and lower lambda to keep EGT in check.

If this is not the case, then you are trying to solve an issue with tuning that was made by your super duper engine builder, who built the engine with too high compression ratio for the required boost.
This causes you to run too late timing and your EGT problems.

You are not going to solve mistakes in hardware with software mapping. Take the heads off, install compression dropping headgaskets, and your problem will magically be solved.

Instead of further theorycraft start posting logs. You have hardware issues or huge issues with mapping to cause high EGT. The RS4 does not have EGT problems.

Where / how is the minimum lambda target value defined when TABGSS is active?

LAMBTS

https://s4wiki.com/wiki/Tuning#LAMBTS

Thanks for your link. Unfortunately, the LAMBTS section does not explain what effect exceeding the EGT of TABGSS has on the variable lambts.

Therefore, I would like to show my understanding problem with an example based on the original K-Box software.

Example
Engine is running constantly under full throttle in sixth gear for 30 seconds at 6800 rpm with a relative fill of 160%. The temperature at the beginning of the full throttle run is 760 degrees Celsius and after 30 seconds of full throttle the exhaust gas temperature is 950 degrees Celsius. Then, according to my understanding, the target Lambda value after the 30 seconds full load run is as follows:




Up to this point, TABGSS has not taken into account because EGT is lower than 979.99 degrees Celsius.

Assuming the full throttle run in our example lasts another 30 seconds and the EGT rises to 1000 degrees. My understanding is that the target lambda value remains at 0.6651 as long as EGT is below 979.99 degrees Celsius.

But what happens to the target lambda value once the EGT reaches 979.99 degrees (threshold from TABGSS) and the EGT rises to 1000 degrees?

Data from 8D0 907 551 K

LAMBTS and ATR are two completely different modules.
They are not connected to each other, output is minimum always.

If you have 0.6 lambda in BTS you need to learn to tune.

Again, post your logs, there is no point to theory craft.

I didn't know that, thank you for your appreciated advice.

The lowest possible value of the ATR module is lambda 0.6?


Yes, that's why I'm here, because I want to learn from you and other experienced tuners how to tune a ME7.

Technical details
- Car: S4 B5
- ECU: D-Box
- Modifications: all stock except precats removed

Log details
- Full throttle run with 4th gear.

Comments
Over the weekend I worked on the tuning. I reduced the boost pressure in the upper rpm range. Afterwards the EGT always stayed below 945 degrees. Unfortunately I don't have a logfile of this, because the software "Visual ME7 Logger" didn't run stable (graphical user interface "freezes" again and again. Sometimes after a few seconds or 2-3 minutes). I have to get "Visual ME7 Logger" running stable first.

What your D-box 512 kb ECU have with RS4 file? ???

551D is AM29F800BB

Log looks quite decent, but you could run it richer already in midrange. 0.8-0.81 lambda instead of what you have now.
It does not hurt power. But not super important.

Also, your intercooler is quite done, you can see 50C IAT at end.

If you want to run this amount of boost, configure the overboost function and make it be active for max 10 seconds.
Set TVLDOBSO also a little higher than 1 second, maybe 2 seconds or so, so it does not get reset on gear shift.

And then TLDOBN has to be set based on when the EGT has "cooled down enough" that overboost should be possible again.
Most likely this value could be set quite high.

Different behaviour in different gears can also be accomplished with FKLDOBG, although as I said before, I don't think this is the solution in this case.
There is no point to start limiting it until the car has been WOT for 10 seconds or so.

For logging if you have a J2534 interface you can use my logger, it's free for ME7:
http://nefariousmotorsports.com/forum/index.php?topic=21482.0title= (http://nefariousmotorsports.com/forum/index.php?topic=21482.0title=)
Also you can patch the file to be able to connect even with the ignition turned on using double slow init.
With that it should be stable.

Thanks. First I will give it a try to use the trick
I read in
http://nefariousmotorsports.com/forum/index.php?topic=837.msg127360#msg127360
and in
http://nefariousmotorsports.com/forum/index.php?topic=11022.msg97250#msg97250

Maybe the connection become stable with that change. I already use a separate K-Line connection to the ECU.

The lowest possible value of the ATR module is lambda 0.6?

Does anyone know if the lowest possible value of the ATR module is lambda 0.6?