Announcement

Collapse
No announcement yet.

XDF/ADX development

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • XDF/ADX development

    Just a heads up dimitri, i found an error in the XDF file. (Ignition - P/T, no knock - meth failsafe triggered) map and (VE table) map are pointing to the same address which is 0xA074.
    Last edited by vwnut8392; 16th December 2017, 21:30.
    "The really good drivers got the bugs on the side windows." Walter Röhrl

  • #2
    Originally posted by vwnut8392 View Post
    Just a heads up dimitri, i found an error in the XDF file. (Ignition - P/T, no knock - meth failsafe triggered) map and (VE table) map are pointing to the same address which is 0xA074.
    Can just delete it. The meth failsafe stuff is not used, it is an old leftover from legacy. "Race fuel" maps are used only. That said correct way is to fork then fix and submit a pull request.

    Knock control works perfectly with aftermarket in my experience.
    http://tuner.ee - http://www.facebook.com/tuner.ee

    Comment


    • #3
      Check your messages dimitri, i updated the XDF for you. added descriptions from the wiki, made all of the important maps have color on the 2D tables, and a bunch more stuff. XDF is in your inbox on here with a more in depth description of what all i changed.
      "The really good drivers got the bugs on the side windows." Walter Röhrl

      Comment


      • #4
        I was bored today with the crappy weather here so i decided to make a few changes to the motorsport features to suit my needs and i figured i'd share them with everyone. i re-wrote how the launch control and NLS works. in my version here you ground pin 39 to the ECU instead of giving 12V. it still retains dimitri's original knock control disable code as well. lastly i added a new feature to the setup and now the check engine light doubles as a shift light! this feature does not harm the real CEL so if there is a fault the shift light is over ridden and the CEL will stay on.there my be a bug with the shift light code that make the CEL flicker, my car did it but thats the only car it was tested in. it may not be useful to everyone but its something else added to the bag of tricks the motronic can do. the base concept of my setup is the same when it comes to the ignition angle degree setup and which feature uses or shares with which one. the launch control uses its own ignition angle variable and the NLS/hard RPM limiter share the same ignition angle degree variable. with my setup also you must have the clutch pressed in or launch control will not activate, this makes it a little more stealth but it is still based off of wheel speed too.

        The XDF is a modified version i made for me. one thing i changed besides reorganizing is i redid all the boost based equations for the boost chip and made them in PSI relative to vacuum instead of KPA. for example the MAP target table cell values now displays in PSI relative to vacuum instead of KPA. i setup the color coding of all of the major 2D maps and added descriptions from the wiki so if you mouse over a value it will tell you whats up lol. you must use this XDF is you want to use this motor chip binary.

        The file attached looks like a text file. the forum will not allow me to upload XDF, BIN, or RAR files. all you have to do is rename the file from PRJmod_DiffLC_ShiftLight.txt to PRJmod_DiffLC_ShiftLight.rar than you can unpack the XDF and BIN file. i look forward to doing a lot more with this and i hope others out there do too! cheers guys!
        Attached Files
        "The really good drivers got the bugs on the side windows." Walter Röhrl

        Comment


        • #5
          Thanks. Interesting additions have you added them on a branch on GitHub Too?
          Greg

          S2Forum.com Administrator & Webmaster

          '93 Coupe with a few tweeks

          Comment


          • #6
            Attaching stuff here is pointless. Make a branch and github pull request.
            Also, PSI/kpa conversion is pointless too. It can be kpa or mbar, but I see no reason to use US units - if you want to, you can change it for yourself, but that will never be committed to the main repo.

            NLS/LC, I have to check the board, but IIRC it was an ADC which without any signal was 0.
            http://tuner.ee - http://www.facebook.com/tuner.ee

            Comment


            • #7
              Originally posted by Greg_S View Post
              Thanks. Interesting additions have you added them on a branch on GitHub Too?
              I dont have github.



              Originally posted by prj View Post
              Attaching stuff here is pointless. Make a branch and github pull request.
              Also, PSI/kpa conversion is pointless too. It can be kpa or mbar, but I see no reason to use US units - if you want to, you can change it for yourself, but that will never be committed to the main repo.

              NLS/LC, I have to check the board, but IIRC it was an ADC which without any signal was 0.
              eh, as i said in my post i changed up the XDF for my personal use, i live in the US and use US units of measure. im fine if with it not being committed, i just wanted to share some work.im pretty sure pin 39 was a grounding input from the factory, its the one i used a long time ago to swap the timing maps when WOT BOX was enabled for launch control. maybe i did change something on the board to make it work years ago and forgot that i did it but im pretty sure there was no changes made back than.

              I thought about making a PDF for you that has different equations in it that if people want to change the units of measure they can just copy and paste from the PDF to get values to display. i always find this to be one of my pet peeves when i work with any tuning software. some developers like boost in KPA, others like it in BAR, and other like it in PSI so why not give the option to have all 3? or with the AFR, some prefer 14.7 AFR and others like lambda. i was happy when i seen you had setup fueling in AFR instead of lambda.

              "The really good drivers got the bugs on the side windows." Walter Röhrl

              Comment


              • #8
                vwnut8392, What's the point of referencing Absolute pressure MAP sensor as relative to vacuum? The vacuum needs external pressure that is measured with F96, and we have correction maps for that (if used). You really can't say a fact that external absolute pressure (altitude pressure) is constant. Now with this assumption you ruin the whole idea of MAP sensor and you really can't tune MAP sensor maps with such irrelevant correction.

                If for some reason the boost control was based on relative pressure sensor, then you won't need that correction anyway.

                Comment


                • #9
                  Originally posted by kymi View Post
                  vwnut8392, What's the point of referencing Absolute pressure MAP sensor as relative to vacuum? The vacuum needs external pressure that is measured with F96, and we have correction maps for that (if used). You really can't say a fact that external absolute pressure (altitude pressure) is constant. Now with this assumption you ruin the whole idea of MAP sensor and you really can't tune MAP sensor maps with such irrelevant correction.

                  If for some reason the boost control was based on relative pressure sensor, then you won't need that correction anyway.
                  i know vacuum has nothing to do with the boost mapping but boost in reference to vacuum is what im accustomed to. its a unit of measure, there is no actual set in stone unit of measure for pressure. its all in how that person at that given point in time thought pressure should be measured. i grew up measuring boost in reference to vacuum because thats how its done in the states for the most part. you go anywhere and buy a boost gauge it has vacuum and PSI. finding a boost gauge thats in BAR or KPA in the states is a hard thing to do when looking for an off the shelf gauge in a store.

                  but my conversion is 1-1 with the original KPA equation dimitri plugged into tuner pro if you look at it. the equation is still the original actually so the cell values are calculated as KPA first than i added on more to the equation to convert it to PSI relative to vacuum. lets take the cell value of 12psi at 7500RPM and 112% TPS on dimitri's stock boost chip BIN. like i said, my XDF is showing 12psi relative to vacuum, now if i remove my math from the equation it comes back to 181KPA which is 100% the same value if converted in either direction. i simply made it so that people in the USA who prefer their units to not be in KPA or BAR to have the option of viewing the table with values we are more comfortable with. not trying to raise an argument, just adapting the XDF to units of measure that i understand more easily.
                  "The really good drivers got the bugs on the side windows." Walter Röhrl

                  Comment


                  • #10
                    I have made some headway into using the data logging in tunerproRT directly. got most of the data logging variables functioning and setup a virtual dash, linked some of the most important tables with the ADX for live tracing of the tables via the ADX. here's some screen grabs of the ADX so far. there is still testing that needs to be done on it yet as i have not driven my car with PRJmod yet because of lack of wideband. im waiting on my 14point7 spartan 2 OEM wideband controller to come in than i will test the 0-5V wideband input i added to the logging. here's some screen grabs of the ADX so far. i know not everyone like boost measured in PSI relative to atmospheric pressure so i will be making another XDF that has boost in KPA and wheel speed in km/h for you european users. i'd like it to be friendly with everyone.

                    ADX 02.JPGADX 03.JPGADX 01.JPG
                    "The really good drivers got the bugs on the side windows." Walter Röhrl

                    Comment


                    • #11
                      As of right now i have a strange glitching problem in tuner pro logging. i know its not the ECU code as when i use winlog with PRJ's stuff to compare his moves data smoothly without an hiccups. i will need to see if there are some com port settings in tuner pro that i have botched up to get it working more smoothly.

                      The 0-5V wideband input on ECU pin 46 works for the most part as well. i have it setup to use a lookup table in the ADX. 0 DEC = 10.00AFR 120 DEC = 14.70AFR and 255DEC= 20.00AFR. this lookup table can be easily changed to suit anyones 0-5V input if it doesnt match. my lookup table is based on the chart from the AEM uego user manual. i will be doing more testing with another wideband controller from 14point7 as soon as i get it.
                      "The really good drivers got the bugs on the side windows." Walter Röhrl

                      Comment


                      • #12
                        It is easily possible to add a loggable ADC and hijack MAF_C or MAF_T for it (since they are unused right now in the stream), but that means moving either the NLS or the Mapswitch logic to another pin, which is totally possible, but I don't have time for it right now.
                        http://tuner.ee - http://www.facebook.com/tuner.ee

                        Comment


                        • #13
                          Originally posted by prj View Post
                          It is easily possible to add a loggable ADC and hijack MAF_C or MAF_T for it (since they are unused right now in the stream), but that means moving either the NLS or the Mapswitch logic to another pin, which is totally possible, but I don't have time for it right now.
                          thats exactly what i did. i think i used the MAF_T but whichever one is at a 16bit value and just pointed it to a new code chunk to look at BE AN 06 and you know the rest. i tested it and it works very well on the map switching input. what im hoping is to move map switching to a pin on the S702 chip instead using the work from BMWman91. that way its all still there and working.
                          "The really good drivers got the bugs on the side windows." Walter Röhrl

                          Comment


                          • #14
                            Getting offtopic here, but there are no 16 bit values, use correct terms.
                            http://tuner.ee - http://www.facebook.com/tuner.ee

                            Comment


                            • #15
                              do you have any suggestions for the glitching issue i have been having? im not sure if its a bug in tuner pro or if there is something that needs adjusted in com port settings. im just running default com port settings at the moment. i have tried 2 different cables, one is the blue cheap KKL cable and the other is the BMW INPA cable which are both just ebay clones. both give the same result.
                              "The really good drivers got the bugs on the side windows." Walter Röhrl

                              Comment

                              Working...
                              X