understanding saab trionic 7 ecu

hello,

my name is Simas and currently i have watched efi tuning fundamentals, understanding afr and practical reflash courses(haven't watched live examples yet) but currently i haven't seen any info on boost control and how to increase air demand. i started tuning on my car which is a saab 9-5 2.3t 2005 136kw from the factory and currently it is running a tune that is supposed to achieve 340nm and 160 kw. This tune was downloaded from trionictunning website which is dedicated to saab tuning. Tuning is made using t7 suite pro which is a powerful open source software dedicated to saabs and it even has live tuning option!

The tune i have downloaded has some parts in the tune that i don't understand. I am attaching screenshot(parts i dont understand are highlighted in red).Boost cal section was also not covered in my watched lessons.Maybe somebody could help me explain those tuned maps that are highlighted and why they are necessary ? I know that they some of those maps adjust how much the wastegate is open during a certain load,but some people in t7 comunity can also achieve higher power just by simply increasing air demand through torque limiters.Is this possible?Also it is worth mentioning that Trionic 7 is a torque/Airmass request system instead of a boost request system.

Second issue i have encountered is negative adaptation. In t7 documentation it is stated that closed loop area's are needed to be tuned first because over time negative adaptation may occur and lean out the fuel mixture through the entire table, but in courses i have seen Andre telling that closed loop areas can be ignored.I am attaching t7 documentation,this is written in page 15 and 16.

The Trionic system is sort of like the Bosch Motronic system. It's possible they even licensed some of the code and control strategies. So if you look up anything for Motronic ME7 or ME17 (mostly VW stuff) you can get some context, although it is not going to be exactly the same.

The way these things are tuned from the factory is that the basic specifications of the engine are decided, and then the engine is tuned on an engine dyno to fill out each cell of the map (or some interpolation is done). Then a bunch of Matlab and Excel tools are used to populate the tables that include the relationship between torque, airmass, wastegate duty cycle, spark, AFR, and throttle.

The various limits are placed on top of that based on requirements for the individual hardware (torque limits to protect transmission, differential, etc for warranty). When we tune the engine in the aftermarket we can't redo the whole calibration from scratch; there isn't enough time or resources to do that. We're going to work around it to get more power, and if possible preserve some of the safety and sophistication of the stock controls.

If you go to page 36 of the PDF you posted, you can see the philosophy of the author for raising boost to get more power. He is recommending that you adjust the relationship between target airmass and the wastegate duty cycle required to achieve that airmass. This implies that you adjust or even deactivate the torque based limits or torque based airflow controls. Page 39 and 40 shows him maxxing out the airmass request, moving you to the right part of the table in page 36. Effectively he is bypassing real torque and airflow based controls by doing that so that boost is controlled by the table in page 36. He's dumbing down the control system.

To address your questions about the tables in your screenshot, I can at least speculate on what these groups do, but since you didn't post the values in the maps it's harder.

1) Aircomp cal - this might be some kind of model based control for the turbo, predicting the throttle inlet pressure. It could be used for a turbo speed model or some other kind of failsafe. I'm about 25% confident about that.

2) Air control - one of those is axis scaling, and the other are probably back-end functions used for the air mass control algorithm. It's hard to say without seeing the tables themselves. Often those function are designed up front by the software engineer, and are set as constants in the ECU that are carried across a variety of engines and vehicles. Basically they're not meant to be changed.

3) Boost cal - go to page 86 of the PDF. Some of these are controls for how responsive the throttle closing is in the event of overboost/too much airflow. Others are miscellaneous things for boost control or air learning and probably should be left alone. The author of the PDF probably doesn't really know what they do or hasn't found a use for them for purposes of making more power.

4) redundant air inlet map - you'll have to look more into it, but it could be an inverse of another table (The Bosch Motronic software has some of this for the airflow control). A lot of times these airflow controls cross check two different calculations to see if there is an overboost or other safety concern.

Talking about adaptation: This is a common concern on reflashing stock ECU's no matter what kind of car it is. Part of it depends on how the closed loop learning is done. A lot of newer ECU's have closed loop learning "zones" based on speed and load (airflow from MAF sensor or some other calculation based on air pressure). That keeps the learning in low load from automatically carrying over to high load. The PDF implies that your ECU does not have that function; if it learns long term fuel trim, it will carry it into all open loop heavy load areas, which could be a problem if you are learning lean.

The general rule of thumb is that you want to keep the short term + long term trims to be within some narrower range, say +/- 5% if you can. There are two ways to do that. One is to disable closed loop fuel, as the author of the PDF suggests. The other is to keep closed loop fuel on and make real time changes to reduce the fuel trims.

In your case I would recommend you set the closed/open loop switching delay to 0 first. If you have bigger injectors, adjust the injector constant per the recommendation of the author of the PDF. Then, if I understand the software correctly, you should adjust the base fuel (engine warm) table to get your fuel trims in line. The key is, adjust with engine warm an closed loop delay at 0. You don't want to be in any kind of warm up enrichment when you are doing this.

Thank you for clearing some things up, i think i will look in to me7 and me17 ecus and how they opreate.as for the boostcal.regmap lets say i set wastegate 100 percent shut through the entire table but if the pedal request map limits airmas to lets say 800 mg/c at 4000 rpm 60 percent pedal pressed than the wastegate would still open acording to the pedal request map(or max allowed airmas map) limits?

The PDF doesn't explain how the torque/airmass limiters decide which actuator to change. Are they all just doing throttle closures, or do they adjust wastegate first? For example on the Bosch Motronic system, there are feedback gains (PID) for the wastegate control to achieve target airflow, and then torque limits that make throttle closures.

You've got a few options here. The first is to try and follow the PDF, even though it doesn't fully explain how everything works and why. The second option is to incrementally raise the torque limits, airflow limits, target airmass, and wastegate table (Boostcal.Regmap) and see if you can get it to gradually raise boost without hitting too many limits.

The third more radical option is to disable all the torque limits (Raise them to a high value you'll never achieve), raise the airflow limit such that it doesn't do anything, and control boost only with the Boostcal.Regmap adjusting duty cycle. It seems possible from reading that PDF, but I'm not sure if you have access to all the tables needed to do that. It's definitely possible on other cars (Motronic system, Subaru system, etc). You just set the wastegate feedback gains to 0 and turn off the overboost/torque limits. Some people find that kind of simple look up table easier to use. After that you could enable just one of the limiters, such as airflow limit, as protection in case of overboost.

at documentation page 22 it gives some hint how torque limiters operate and in step 13 it says that the throtle requested mg/c is compared to current MAF mg/c and the actuator is adjusted acordingly but to what extent it is not known. I may experiment with stock turbo maps but without a dyno it would be hard for me to tell if the boostcal.regmap gets overided by other limiters in order to open the wastegate.I prefer the second method of gradually increasing limiters. I guess there is no "correct" way to tune as long as the car is making target power and the afr,ignition is set correctly?Currently the car runs fine on this tune, i experience an overshoot from time to time but nothing dramatic, so i will leave this tune as it is. My main goal now is to understand what makes a good stage 1 tune and work my way up from there :).when i made my tune based purely on torque limiters i was told in the forum that it would make power but it is not the way to go and told me that my current tune (downloaded from the forum) is a good indicator for stage 1. That's the moment when i decided to turn to hp academy for help because if the car ride's nicely and it is making the target power and the afr,ignition is working well than why fix it...

Make a spreadsheet of the 3 tunes (stock, yours, forum's) comparing important maps. Do a similar WOT pull on all 3. Post the resulting files.

curently i dont have an obd2 cable capable of loging.I have a bdm header soldered on the ecu and the bdm cable is hanging out of the ecu through the front passenger glove box. Flashing is done via usb bdm adapter. I think it is about time for me to purchase obd cable.When i have the results i will let you know.