A little update :
First I asked a friend that is working on Renault-Volvo trucks test bench, the current trucks are running 600°C.
Current diesel cars are running 700-800°C at the exhaust manifold.
Then I though it was interesting to give some informations to people who are interested by this thread. We worked with Stevie to make the better tuning file possible. What we have learned :
The ECU of om642 first generation is EDC16CP31. The calibration file that is contained into Xrom memory isn't fully readable throught OBD port. There is an hidden part that can only be reached by doing a bench tune/opening the ECU. This hidden part contains some copies of maps from the calibration part reachable through OBD port.
The hidden part contains some SVBL (single value boost limit) and copies of boost maps. It's why a lot of people can't get the boost pressure they want with this ECU.
To do a bench tune we used this tool :
[AMT Software Offline Available] ECUHELP ECU Bench Tool Full Version with License MD1 MG1 EDC16 MED9 ECUs No Need Open ECU
The tool can do "service mode", it means that it does not require to open the ecu and can flash the ECU just by connecting some pins. It's possible to fully clone ECU with that tool, and we did it several times.
It seems to be possible to clone every EDC16CP31 with different OEM Mercedes numbers. We tried to write the files of our currents ECU on older units and it worked. It seems that the most important number is the Bosch number : 0 281 01x xxx.
Unfortunately this chinese tool isn't reliable and I encountered some writing faults, especially during checksum block writing despite the checksum were verified by Winols and the software of the chinese tool.
The problem seems to be software related, because sometimes deleting the software and reinstalling it get the tool working again. Sometimes not. Sometimes it's needed to change computer. And sometimes your ECU got killed by the writing error (2 times for me, so 2 ecu killed).
So if you want to remap with this tool, be carefull because you can get problems. I won't use it anymore in the future.
Now about what we did to remap Stevie's car and mine :
The goal was to run a fully calibrated tuning file. It means no tricking the ECU to get a higher injected quantity or boost pressure. To do that we had to search, find, and modify all the limitations of the ECU. The advantage of doing that is that we can respect all securities of the file, extend the maps using straight line and polynomial equations, running calibrated smoke maps. Actually we wanted an OEM quality file. And we succeeded. Now we can overcome the 100mm3 maximum injection of stock configuration, and put the boost and rail pressure we want without any limitations.
The second problem was to datalog the car. It doesn't exist any good software for Mercedes as for VAG or BMW, and xentry isn't great for tuning purpose. Then I succeed to understand how display all variables I want in charts with vediamo. We don't have any logging limitation value as the common boost pressure is limited at 2650mbar by CAN. Actually we are logging directly the value the ECU is using. We succeed to display more that 3200mbar on the datalog without changing the can offset and coefficient. It was difficult for me to understand how it works, but Vediamo is a very powerful tool. A look into how looks the graphs :
Here, measured boost pressure was 3118 hPa while asked boost was 3083 hPa at 3500rpm and 101mm3 injection.
The backpressure was 3931 hPa :
Anyway, don't really look at the values on the charts, it's just one example of the multiple trials we did to find a proper tune.
The datalog was helpful to look after VNT behavior, but I will explain more things about the turbo after.
As Mercedes ECU smoke maps are based on MAP sensor pressure regulation contrary to MAF sensor and Lambda sensor as BMW and VAG, Stevie wanted to try a MAF less tuning file. This way, any future MAF sensor problem can't occur.
To do that, I switched the MAF sensor presence to 0, and enabled ASmod modele calculation. With the help of another tuner, I succeed to delete the DTC to totally unplug the MAF sensors.
The result was surprising. The car is running much more smoother and the gear pass smother too. So Stevie physically deleted the MAF sensor from the boost pipe.
Now, about how we tune the car, the goal was to get a safe file and without visible smoke at the exhaut. The goal wasn't to maximise everything as Stevie is often pushing hard the car on the autobahn. The maximum we tried was 750nm and 330hp calculated, with 2,1 bar boost pressure to avoid smoke and trying to run low exhaust temp. We could not go higher than 330hp and actually it wasn't a safe file neither.
The problems of om642 are exhaust temp and cylinder pressure. The pre-turbine temperature sensor Stevie installed was helpful to set the proper settings. To decrease exhaust temp, either you rise the boost pressure to lower the AFR, either you add some SOI degrees more. But this two solutions are increasing the cylinder pressure.
As some thread about om642 cracking head pistons were published, we were afraid to be too violent with the engine. So it's why we were careful when adding SOI.
I calculated how many degrees were needed to keep the same End of Injection (EOI) as stock 80mm3 with 90mm3 and 100mm3. The results were respectively 2° and 4° more just to keep the same EOI. We only tried 2° more as why think 4° are already a big deal for the engine already running 2,1 bar relative boost pressure. Maybe we are wrong but as we don't have any informations more, we didn't want to be the first to discover the maximum SOI you can put before blowing up the engine.
So the car was running quite well, but the exhaust temp was too high, approx 900°C. It's obvious we had to put more SOI to get conservative exhaust temp. But we did not. Then I will explain my vision :
OM642 is made to be low rev torque machine and the stock power is at 3800 rpm. The engine isn't made to do such great power as BMW and VAG. Because it's designed to do torque. A lot of torque. It's why I think it would be less dangerous to have a 800nm with 300-310hp and a very beautiful usable rev range, than a 330-340hp. If you compare with other german engines, om642 is the easiest to get massive torque.
So I met Stevie in june 2024 as I had to go to Berlin. I tried his car I tuned. The power was great but the turbocharger is too big. YES, it is. Don't listen to hybrid turbo makers and others guys who want to sell their products. My feeling when driving the car was mixed. Indeed when pushing on the autobahn the car is really fast. But the reactivity is poor. When pressing the pedal to get a full load, you have almost 2 seconds to wait before the ECU hit the boost limiter, then the gearbox shift down to rise the RPM and give the turbine the necessary airflow. 2 seconds it's enormous. Believe me, you don't want it. And it's just a 2365 turbo. Not a 2573 or something else bigger. Indeed we tried to close the VNT and run richer mixture to get faster spool. But impossible to make it spool before 2200rpm. As the peak power is 3800rpm, the range usable isn't that great. Understand that under 2200rpm you won't have any torque. This engine isn't like BMW or VAG that can run 4500-5000rpm. OM642 can make 750nm at 1800rpm. It's a totally different feeling when driving a beast torque than a poor laggy engine. A lot of low torque is a very different feeling than a high rpm power engine.
So the decision is made. As it is difficult to go higher than 310-320 hp without getting to high exhaust temp or rising SOI above limits we don't want to go up, Stevie will change his turbocharger to get Ballbearing 2260vklr from OM642LS. There are plenty in the scarpyard and you can replace the internals for approx 300 euros. Really less than a 1700 euros turbo that is spooling late.
The 2260VKLR can deliver 2600mbar at 1600rpm and 2700mbar at 3600rpm in the stock LS software. It will be easy to get 2800mbar at 1800rpm and 2900mbar at 3600rpm. The maximum Stevie is running now is 2,1 bar boost. 1,9 bar with the 2260VKLR isn't very less. But the usable range and the reactivity of the turbo will transform the engine.
Hope those informations would help some of you to take the decision to choose or not an hybrid turbo. I choosed the 2260vklr way. I don't regret it. And I'm sure that Stevie won't regret it as well.
If you have any questions, or if you need any tuning on OM642, I'm now able to do it. We'll try to keep the thread updated.
34plus2.jimdofree.com
