Playing around with the 2.7 software (ECU and TCU)

[JeanLuke]

I think the biggest issue here is that you have run way ahead of (almost) everybody else in this forum. I mean that as a complement. The consequence is that, having exceeded the knowledge of most here, you are going to have to cut your own path.

That said, I am enjoying watching your posts as you figure it out. Please keep going.

Regarding the TCU, I did see a project on the internet a few years ago where somebody built one out of either an Arduino or the native AVR chips that power the Arduino (or something similar). I did a quick Google but could not find it again. They had it set up with an LCD and paddle shifts. Pretty cool stuff.

[RTB]

Thanks JeanLuke. I had been wondering if I should keep posting my journey or not. Wasn't sure if people were benefiting, or it was just getting annoying. Even for those few who are no doubt waiting to see something fail miserably (Here's hoping I never make this group happy).

I have seen an aftermarket TCU built on the Arduino base and fully plug and play. Being able to play that much would be nice, but really. the EGS52 has all the hardware required and has spare I/O. Just having the knowledge to modify the base code could open all of this up. Guess it is more about the challenge of the journey than the end result.

Thanks back to you also for helping broaden my vision when looking at this stuff.

[RTB]

As for the paddle shift option. I can't see why that isn't at least possible to some degree now.

As far as I can tell, our cruise control is managed by the ECU. If the speed goes too high under cruise, then the TCU will drop back a gear to help maintain the set speed.

There is no hard-wire signal from the ECU to TCU for this, so it must be done via comms. There is a max gear variable in the TCU. At this point I assume that the ECU sends this max gear signal to the TCU to control this kickdown.

I made a OBD sniffer in Excel a while ago. I keep meaning to use it to record the comms messages flying around and see if I can nut out the kickdown signal. Then, all we would need is a micro-controller to replicate this signal (I like the ESP8266 running Lua for simple tasks). Wouldn't mean we could change up beyond what the base TCU map asks for, but would allow a basic paddle shift style system that could help to easily hold a gear. My TCU is tuned to rev reasonably low, so would be beneficial for me since we don't have a sports mode button. Not that moving the shifter is that hard, but again it is more about the challenge. The longer I have the Jeep, the closer I get to committing the time to look into this one.

Edit:- In my mind I would use the radio volume up/down buttons on the steering wheel for this. I have already recorded their commands using the Excel sheet. The Microcontroller could just listen for the volume up/down and send the gear up down command in response. I have an aftermarket radio, so these buttons are unused now.

[JoeTe]

Originally Posted by RTB

Mainly just looking to learn really. That said, the more I learn, the more I can see some potential to optimise.

What I really wish I could look into more is the actual program that runs it all. Not so much on the ECU, but for the TCU. I have a mate that loves his Fords. He is able to modify the full structure of the running code for his TCU. I am yet to find anything that allows that for the EGS52 fitted to our cars.

Strange below with the C30 EOI that it advances and retards as the revs go up. I need to look more into diesel tuning. The internet just does what it is best known for, contradicts itself with every new page read.

As I play, I have started putting together SOI and rail pressure mods to fit to my philosophy of what the SOI - EOI should look like. This again is just for learning right now. I mean it isn't new stuff. Sure, every engine is different, but someone out there must know the basics for what is best as a general rule.

I have seen comments like advancing the timing will increase power, but reduce economy. But if I increase power at any given point, I will then need to reduce fuel to maintain the original power, thus improving my economy???

The opposite comment I have read that by retarding timing the power goes down, but economy improves. Well if I am on the fwy, and I retard the timing which then reduces power, I will slow down. This means more throttle, more fuel for the same load????

What does make sense is that if the SOI goes further into the compression stroke, then this takes away power due to increasing cylinder pressures while the piston is still trying to compress the gases (reverse torque). But at the same time, if the EOI goes further down the power stroke, then we take away peak combustion pressures and the power goes down. So ideally we would start and stop injection at TDC, or a tiny bit before as RPM increase (to allow for the combustion time delay). Doing this though would see extreme combustion temperatures and pressures and burn holes in pistons and wipe out bearings. Not to mention requiring rail pressures extremely high and Nox emissions through the roof. No smoke though!!

Since a crank angle of less that TDC +- 10 deg really has minimal effect on torque, it would seem reasonable that at light load/RPM the SOI should be just before TDC, and the EOI could go anywhere up to around 10 deg ATDC. From that point the SOI would need to advance at a rate slightly faster that the EOI moves back as injection duration increases. This would create the average injection timing advance as IQ and RPM increases to allow that little bit more time for combustion to complete.

A knock sensor, or at very least a knock prediction calculator would obviously be very helpful for engine health.

Am I way off the mark with this??

I think you are in the right way. SOI map in our engine and c220 are the same but c30 is different and have more SOI than ours so I guess will be based in more performance and our can be based in more fuel consuptiom. Then my conclusion is around EOI 5º degrees should be the best for performance but when you going up you have a lot duration so your EOI will be increased for durations reasons reaching until 10 degress its still safe or less agressive temperature for engine.
Anyways a was speaking with a guy that make a test with EOI 0º- 5º - 10º- and although he didnt use a dyno he only founds difference in engine sound but performance measuring rpm with throttle was the same. People thinks SOi is ignition timing in petrols engines and really is very similar but probably 2-5 degrees in petrol engine have too much difference than SOI in diesel engine. Probably a little bit power will should be change but dont feel so in my opinion SOI map should be adjusted for reducing EOI at high IQ and if you want in EGR zone if you have off

I think when people say that you will have better fuel consuptiom with less power (same theory that mod n75 for fuel consuptiom) can be better because aceleration will be slower and here is where you save fuel but your engine is not more efficiency but if when you have to acelerate you use 30hp and other use 50hp in that time you will consuptiom more but if your engine is more efficiency (and probably if you adjust EOI you can do it) if both use same power all time your consuptiom will lower.

Brands of course always want best fuel consuptiom, efficiency and reability so I guess STOCK SOI should be very fine only emission will be afected to dont have the best SOI. Is interesting you can download 1.9 TDI 1998 and see the SOI and you can see how is not the same than other 1.9 TDI with more agressive emissions system

Anyway is a very interesting thing so i would like you keep posting yours advances

Originally Posted by RTB

Thanks JeanLuke. I had been wondering if I should keep posting my journey or not. Wasn't sure if people were benefiting, or it was just getting annoying. Even for those few who are no doubt waiting to see something fail miserably (Here's hoping I never make this group happy).

I have seen an aftermarket TCU built on the Arduino base and fully plug and play. Being able to play that much would be nice, but really. the EGS52 has all the hardware required and has spare I/O. Just having the knowledge to modify the base code could open all of this up. Guess it is more about the challenge of the journey than the end result.

Thanks back to you also for helping broaden my vision when looking at this stuff.

About EGS52 is very interesting thing. Mercedes with EGS52 have M-mode and other more and i guess we can activate it too but dont know how we can program some buttons for this

http://blog.obdii365.com/2022/07/21/...722-9-gearbox/


Pd: Sorry for my english im not very good explain in this language

[RTB]

Hi JoeTe,

I wonder if the C30 uses higher flow injectors and that is part of the reason why their SOI moves forward a bit? Really need to see the duration maps too as they do work together. I also wonder if the difference your mate heard with the 0 - 5 - 10 deg EOI was Knocking?? An EOI right at TDC 0 deg seems very brave to me. That means all injection, combustion, and pressure rise is happening during the compression cycle while the piston is still moving up. That must put some strain on the crank and conrods.

I think people who compare diesel and petrol tuning might not be looking at it correctly. In our diesels, we get a nice long burn that "normally" should last a similar time to the duration of injection. So with the fuel rail pressure map, we can control how fast or slow our burn happens, how long the burn takes. In a petrol, the burn happens in an semi-uncontrollably fast way right after the spark plug fires.

With the M mode in the gearbox. We do have 3 different gear change schedules in out TCU. My understanding is one is for high range (the one I play with), 1 for low range, and the other is unused.

I know for sure that input pin 3 on the TCU is the summer/winter switch on one of the Mercs, but for us it is used for one of the gear selector positions from the shift lever. I never really use low range, so I have toyed with the idea of just having a switch in the car with the correct resistors in it to mimic the transfer case switch. Change the low range ratio in the TCU to 1:1 to avoid any ABS sensor faults. Then, I could change the switch position from high range to low range while keeping the transfer case in high range. This should map select both different change maps in the TCU and different Drivers wish maps in the ECU. So sports/economy mode. What the downside would be that if ever I actually put it into low range, then the TCU would now see a ratio fault and go into limp mode. This is the main driver for me want to be able to modify the TCU running code. Then all problems would go away and I could have the best of both worlds.

As for the state of play with SOI. I am back to needing to do more research.

Turns out there is a big difference with SOI when converted using a time based to crank angle conversion, or using the fixed multiplier mentioned earlier. What is weird though is that the results produced in much of the map is similar, but the centre right is way way different. Way too different for my comfort.

Time based on the left, fixed multiplier on the right.



The differences between them.

[RTB]

Learnt something about the swirl flaps today.

I have stopped looking at the SOI for a bit as I do more research. The maps in our WJ EDC15C2 are different to many others, and most the info on the internet is for other vehicle types. Each manufacturer puts their maps in different locations, etc, so makes it a pain.

Anyway, found out that the swirl flaps have some sort of link to inlet air temp. It is 36 deg C here today. I was watching the swirl flaps on WJDiag Pro. They were doing their thing for a while after I started the car, then suddenly they went full open and stayed there. Came back, after the car cooled down went out again, and after a few minutes the same thing happened.

So it would appear that if the inlet temp goes high, the swirl flaps are disabled full open.

[OffroadSpike]

Makes sense, more temperature = less air density, so it needs more air volume for the same mass