Page 1 of 7 12345 ... LastLast
Results 1 to 20 of 124

Thread: 2.0TFSI Engine build + HPtuners?

  1. #1
    Tuner
    Join Date
    Dec 2019
    Location
    Alingsas, Sweden
    Posts
    66

    2.0TFSI Engine build + HPtuners?

    Hi everyone,
    I have a 2010 Audi A4 B8 2.0TFSI (EA888 gen2, CDNC) Avant Quattro with a unknown stage1 tune.
    When i hooked up the MPVI2 and tried to read the tune a got a bunch of popups saying that the control module is not readable and definition failed to load. I am not able to view anything in the tune file which is created.
    From reading up on the forum I've understood that this might be common with these ECU's. I did open a ticket for this but i have some thoughts I would like to run through you meanwhile.

    I am completely new to HPtuners and MED17 but I have some experience and understanding from engine development of larger engines, mostly diesel though.

    The stock tune, supplied by HPtuners, that I might have to start from is that factory equivalent with full control functionality?

    I would ideally like to be able to open the existing stage1 tune and compare that to the stock tune to understand what parameters that have already been changed.
    I do however think I'm starting to get a grip on what to change from all of you other contributors here on the forum, THANK YOU!

    The existing stage1 tune is supposedly raising the power output to 270Hp (according to the seller of the car).
    Will I with HPtuners be able to produce a tune performing similar to a stage 1 tune? I'm just a bit hesitant to trow away a tune I know works OK.

    Would it by a lot of logging be possible to somewhat understand how the stage 1 tune is set up even though the tune is not readable?
    If someone would like to share a logging channel config file suitable for these engines I would be super grateful.

    Just to give you a background on what I am trying to accomplish:
    I bought the car 18 months ago. Two days after I bought it I lost oil pressure resulting in a complete engine rebuild. The mesh screens in the lubrication system had broken from fatigue.
    Small pieces of the mesh screens traveled with the oil causing the balance shafts to seize and the piston cooling nozzles to be stuck open resulting in low oil pressure on idle.
    I ended up putting in forged pistons and rods and deleted the balance shafts. Deleting the balance shafts meant i had to reconfigure the lubrication system and blocking some of the oil passages in the block.
    Deleting the balance shafts also meant I needed to go for a electric water pump since one of the balance shafts are driving the mechanical water pump in the original setup.
    The electric water pump is controlled by a standalone micro-controller and is right now only dependent on coolant temperature but i intend to include throttle position or maybe a load signal in order to keep the temperature peaks down.
    I have also fitted a 034-downpipe with high flow cat, a open filter and a larger throttle body.
    The intention is swap out the turbo in the coming months. To combat any fueling limitations I have also upgraded the low pressure fuel pump to a Walbro 525 and tweaked the fuel pump control module.
    On the injection side I have mounted the intake manifold from a B9 that has port fuel injection as well. The PFI injectors will most likely be controlled by a piggyback system.
    Ideally this functionality should have been included in the MED17 controls but I don't have high hopes for that.
    Swapping out the intake manifold resulted in a different scaling of the runner flap potentiometer. Would it be possible to change the boundaries of this sensor with HPtuners or should I just switch of the diagnostics?
    I also intend to incorporate a flex fuel system but I am not sure how to do the controls of this as of now.
    I know there are a flex fuel version of this generation of 2.0TFSI but I don't know if its possible to include this functionality in the factory ECU with HPtuners?

    Extremely grateful for any answers or comments to my many questions and thoughts above.

    And oh yeah, did I mention that this is my daily that i take the kids to school in and drive 100km to work each day

  2. #2
    Tuner in Training
    Join Date
    Dec 2018
    Posts
    46
    Cool setup, sounds very similar to mine. I'm curious how you configured your cooling system. I ended up using an EWP150 and one of those aluminum water pump housings with some of the passages welded so that I could pump water in externally without the mechanical impeller blocking flow.

    Regarding reading the tune, HPT probably just has to add support for your vehicle. Once they get it working, VCM Editor will read the binary that's currently on your ECU. If that works, then you should be able to view the current stage 1 tune in VCM Editor. You may have to ask others for a stock file to compare against. You can probably also post your stage 1 read here and people can help you figure out what has changed.

    There is a chance that the stage 1 tune you have includes a "lock" that will prevent you from reading it out via OBD. If that's the case, you might try contacting the previous owner to see if you can figure out how it was tuned and if it can be flashed back to stock (e.g., through the workshop that originally tuned it). Alternatively, if you can get a stock read for your car somehow you can ask HPT to wrap it into an .hpt, if it isn't already, and do a "write entire."

    Your problem is probably going to be figuring out what the stock values are because you'll be starting with the stage 1 read, but to answer your original question, you can probably reproduce a stage 1 tune pretty easily. You can understand a bit about the tune by logging load, timing, etc. but that won't tell you everything. Tables interact in various ways affecting the final load, timing, etc. that the ECU runs with. There are ways to log variables in memory, but they're beyond the abilities of most available tools.

  3. #3
    Tuner
    Join Date
    Dec 2019
    Location
    Alingsas, Sweden
    Posts
    66
    @mukhery Thanks for your reply, seems promising then
    I do have access to another car, same model, same year, same engine, same trans, even the same color but with a stock tune, maybe that can be to some help.

    Regarding the cooling I tried a bunch of different ways to get the plastic water pump housing to work but I ended up, just like you, getting the aluminium one, machined it so there was basically just the thermostat housing and the flange to the block left and then made two new large hose barbs to the electric pump. I went for a Pierburg CWA400 which I am controlling with an Arduino that is reading a bunch of stuff over CAN. I will probably end up redoing the communication part though so I don't have to switch it off every time I am logging or flashing a new calibration. But other than that and that it is a bit hard to bleed, I have to admit that I am quite satisfied with how well the cooling system is performing

  4. #4
    Tuner
    Join Date
    Dec 2019
    Location
    Alingsas, Sweden
    Posts
    66

    First log!

    Update: HPTuners managed to add support for my vehicle and I am now able to read the tune which is currently in the ECU. The last week I have been messing about to improve the controls of my coolant pump. I wanted to get away from reading the temperature over CAN and improve the PID controller. I finally managed to get it working to a point where I felt safe to do my very first log with WCM Scanner!
    The temperature still spikes a couple of degrees during a pull but I have some ideas on how to improve this.

    Anybody who likes to look through my log and tune file and give any pointers are more than welcome.
    Attached Files Attached Files

  5. #5
    Advanced Tuner
    Join Date
    Sep 2019
    Location
    ON, Canada
    Posts
    346
    Quote Originally Posted by AlexOlen View Post
    Update: HPTuners managed to add support for my vehicle and I am now able to read the tune which is currently in the ECU. The last week I have been messing about to improve the controls of my coolant pump. I wanted to get away from reading the temperature over CAN and improve the PID controller. I finally managed to get it working to a point where I felt safe to do my very first log with WCM Scanner!
    The temperature still spikes a couple of degrees during a pull but I have some ideas on how to improve this.

    Anybody who likes to look through my log and tune file and give any pointers are more than welcome.
    do you have a boost gauge?

    i don't think this file is a tuned file as most of the tables are still stock, you may want to check for boost leaks because your car is reading load that's too high and its causing throttle cut as well as a slightly rich condition

    TC-Tune.png

  6. #6
    Tuner
    Join Date
    Dec 2019
    Location
    Alingsas, Sweden
    Posts
    66
    No I don't have a separate boost gauge at the moment. To verify the TIP or to check for leaks?
    Would a stock tune be running 2.6bar TIP with a high flow cat and replacement air filter?
    You might be correct on the boost leak, I think my PCV is leaking slightly and my modified intake manifold might still be leaking slightly. I have been chasing leaks for a while since switching to the B9 manifold.
    God I have soo many questions and so much to learn, this will be fun

  7. #7
    Advanced Tuner
    Join Date
    Sep 2019
    Location
    ON, Canada
    Posts
    346
    Quote Originally Posted by AlexOlen View Post
    No I don't have a separate boost gauge at the moment. To verify the TIP or to check for leaks?
    Would a stock tune be running 2.6bar TIP with a high flow cat and replacement air filter?
    You might be correct on the boost leak, I think my PCV is leaking slightly and my modified intake manifold might still be leaking slightly. I have been chasing leaks for a while since switching to the B9 manifold.
    God I have soo many questions and so much to learn, this will be fun
    basically my thought process is that because the MAP value you are logging is the calculated map value (not TIP sensor value) it could be a boost leak causing it to read higher than normal at the MAF which in turn cause a higher calculated map value and load overshoots as well as a rich condition (which you have... all together)

    That's just my 2 cents

  8. #8
    Tuner
    Join Date
    Dec 2019
    Location
    Alingsas, Sweden
    Posts
    66
    Yeah ok, @TylerJDubbs you were right. There was a quite bad boost leak right at the throttle. The coupler to the TB have increased in size and didn't seal properly any more. I changed the coupler and it passed my soapy water bubble test, but I don't see much difference in the log. Lambda looks maybe a bit better but actual load is still high during pulls and I managed to get a throttle closure during my first pull again. I did ad the TIP sensor value to the log this time and it matches the calculated MAP value quite well. Because that is the real sensor value right?
    Attached Files Attached Files

  9. #9
    Advanced Tuner
    Join Date
    Sep 2019
    Location
    ON, Canada
    Posts
    346
    Quote Originally Posted by AlexOlen View Post
    Yeah ok, @TylerJDubbs you were right. There was a quite bad boost leak right at the throttle. The coupler to the TB have increased in size and didn't seal properly any more. I changed the coupler and it passed my soapy water bubble test, but I don't see much difference in the log. Lambda looks maybe a bit better but actual load is still high during pulls and I managed to get a throttle closure during my first pull again. I did ad the TIP sensor value to the log this time and it matches the calculated MAP value quite well. Because that is the real sensor value right?
    That's a good find

    seems like it took about 12-15% load off and fueling is closer from fixing that boost leak, but it does seem to run more boost than my car stock but not by much.

    as far as i can tell from your file its stock with a few DTC changes mainly.

    Have you tried tuning it at all?

  10. #10
    Tuner
    Join Date
    Dec 2019
    Location
    Alingsas, Sweden
    Posts
    66
    Quote Originally Posted by TylerJDubbs View Post
    That's a good find

    seems like it took about 12-15% load off and fueling is closer from fixing that boost leak, but it does seem to run more boost than my car stock but not by much.

    as far as i can tell from your file its stock with a few DTC changes mainly.

    Have you tried tuning it at all?
    No not yet. I need to read up a bit more on the torque management before I dare do too much. But I thought I'll start by scaling the throttle body air flow map ECM 35035 to account for the larger throttle I have.
    JNC_Throttle_Comparison.png
    Do you see any issues doing this?

    I would have liked to add rail pressure demand to the scanner as well but its not available. I just want to see that my Walbro 525 isn't the cause of the rich conditions.

    If its not tuned, that's great news, then I can expect so much more!
    Last edited by AlexOlen; 04-28-2020 at 02:37 AM.

  11. #11
    Advanced Tuner
    Join Date
    Sep 2019
    Location
    ON, Canada
    Posts
    346
    Quote Originally Posted by AlexOlen View Post
    No not yet. I need to read up a bit more on the torque management before I dare do too much. But I thought I'll start by scaling the throttle body air flow map ECM 35035 to account for the larger throttle I have.
    JNC_Throttle_Comparison.png
    Do you see any issues doing this?

    I would have liked to add rail pressure demand to the scanner as well but its not available. I just want to see that my Walbro 525 isn't the cause of the rich conditions.

    If its not tuned, that's great news, then I can expect so much more!
    As long as you’re running a properly calibrated MAF without any charge/intake leaks I wouldn’t worry too much about that calculated Throttle airflow map as it’s only really used when the vehicle is operating on the secondary air path (limp mode, with MAF error) but you can ask HP to add calculated airflow as a logging param and tweak the map until it tracks actual MAF airflow closely, this way if your MAF fails it will still drive relatively normal

  12. #12
    Tuner
    Join Date
    Dec 2019
    Location
    Alingsas, Sweden
    Posts
    66
    Quote Originally Posted by TylerJDubbs View Post
    As long as you’re running a properly calibrated MAF without any charge/intake leaks I wouldn’t worry too much about that calculated Throttle airflow map as it’s only really used when the vehicle is operating on the secondary air path (limp mode, with MAF error) but you can ask HP to add calculated airflow as a logging param and tweak the map until it tracks actual MAF airflow closely, this way if your MAF fails it will still drive relatively normal
    Oh OK then that can wait. I was just out to double check what messages I got when reading the existing tune on the car. And surely it's saying that the supplied tune file is a stock tune. I totally misunderstood that before, I thought I was actually able to read the tuned file in the ECU. That explains why you aren't seeing any differences from stock. I thought all differences from comparing to your "B8-A4-2LT6MT-8K2907115AD_0004-Stock" file was due to that mine was tuned.
    A bit scary not being able to revert back to what I have now

  13. #13
    Advanced Tuner
    Join Date
    Sep 2019
    Location
    ON, Canada
    Posts
    346
    As for keeping that tune, from the looks of it it’s not very good anyways (throttle cut, load overshoots, pretty rich on wot for DI) I wouldn’t be too worried about writing over it myself considering your binary is almost the same as mine and you could just copy paste my tune which from logs seems to be better anyways lol

  14. #14
    Tuner
    Join Date
    Dec 2019
    Location
    Alingsas, Sweden
    Posts
    66
    Quote Originally Posted by TylerJDubbs View Post
    As for keeping that tune, from the looks of it it?s not very good anyways (throttle cut, load overshoots, pretty rich on wot for DI) I wouldn?t be too worried about writing over it myself considering your binary is almost the same as mine and you could just copy paste my tune which from logs seems to be better anyways lol
    Haha ok you convinced me. I'll start working on some changes to the stock calibration and post back here in day or two.

    Thanks for the awesome support so far!

  15. #15
    Advanced Tuner
    Join Date
    Sep 2019
    Location
    ON, Canada
    Posts
    346
    Quote Originally Posted by AlexOlen View Post
    Haha ok you convinced me. I'll start working on some changes to the stock calibration and post back here in day or two.

    Thanks for the awesome support so far!
    I got bored, have a look at this

    AlexOlen_B8-A4-2LDSG-8K2907115P_0008-TylerMod.hpt

  16. #16
    Tuner
    Join Date
    Dec 2019
    Location
    Alingsas, Sweden
    Posts
    66
    Quote Originally Posted by TylerJDubbs View Post
    I got bored, have a look at this

    AlexOlen_B8-A4-2LDSG-8K2907115P_0008-TylerMod.hpt
    Haha! you know I was so tempted to just copy and paste your tune changes in to my file, but I stopped my self and thought I would try to properly understand what I was doing before testing it. I guess I will have to learn while testing it

    F*ck I feel so lazy asking all these questions but I'll do it anyways...
    Are the Optimum Engine Torque map values normalized to [ECM] 23279 - Normalized Indicated Torque?
    Why does the engine torque axis on the torque to load only go to 90%?

    And some spark timing questions: Why are we not modifying MBT spark maps and how does base timing to MBT maps correlate?
    And I see you only changed the y-axis for the spark timing and I understand why but, is it safe to run the same timing at 255% engine load as we used to do at 190%? I would have expected it to taper out towards full load, or does the knock control take care of that and we can back off if we see any issues in the logs?
    I'm still an Otto noob...

    Can't wait to try it out! Thanks man!

  17. #17
    Advanced Tuner
    Join Date
    Sep 2019
    Location
    ON, Canada
    Posts
    346
    Quote Originally Posted by AlexOlen View Post
    Haha! you know I was so tempted to just copy and paste your tune changes in to my file, but I stopped my self and thought I would try to properly understand what I was doing before testing it. I guess I will have to learn while testing it

    F*ck I feel so lazy asking all these questions but I'll do it anyways...
    Are the Optimum Engine Torque map values normalized to [ECM] 23279 - Normalized Indicated Torque?
    Why does the engine torque axis on the torque to load only go to 90%?

    And some spark timing questions: Why are we not modifying MBT spark maps and how does base timing to MBT maps correlate?
    And I see you only changed the y-axis for the spark timing and I understand why but, is it safe to run the same timing at 255% engine load as we used to do at 190%? I would have expected it to taper out towards full load, or does the knock control take care of that and we can back off if we see any issues in the logs?
    I'm still an Otto noob...

    Can't wait to try it out! Thanks man!
    honestly I'm not sure about the torque normalization, I would like to think that the torque tables reference of 500nm is the 100% point in its calculation. but that is only an assumption, sounds like you already have a bit of a project

    in most Motronic binaries i have worked with the torque axis of the load table is normally 95-99% not sure whats accomplished with the 90% method

    based on the little bit of Bosch Motronics experience that i have, the stock timing tables are very conservatively mapped above the load ranges we normally enter on a stock tune. i find the best way for starting stock turbo files is to extrapolate the axis to match the new load scale and then adjust timing based on KR.

    MBT timing maps are normally calibrated will beyond what is in our capacity here on the street, so I leave them stock because they utilize fine knock methods to run the best timing possible even on the highest octane fuels. I doubt anyone could do better than the factory in this area

  18. #18
    Tuner in Training
    Join Date
    Dec 2018
    Posts
    46
    Quote Originally Posted by AlexOlen View Post
    Haha ok you convinced me. I'll start working on some changes to the stock calibration and post back here in day or two.

    Thanks for the awesome support so far!
    I think the read you have is definitely not stock. Some of the DTCs are disabled like Tyler said, such as the overboost and underboost codes that tuners typically disable. I think some of the axes are modified too, like the optimum torque monitoring table (ECM 12414) but strangely not the main optimum torque table ECM 32900. I didn't think ECM 12414 was useful for anything other than preventing torque intervention?

    I saw a P2015 code. When you swapped intake manifolds did you remove the intake runner flaps? If so, you'll probably want to disable those codes. There are also some configuration codewords not currently available in HP Tuners that allow you to fully disable the runner flap functionality, maybe the tune you have already has those disabled if you're not seeing a check engine light for the flaps.

  19. #19
    Tuner
    Join Date
    Dec 2019
    Location
    Alingsas, Sweden
    Posts
    66
    Quote Originally Posted by mukhery View Post
    I think the read you have is definitely not stock. Some of the DTCs are disabled like Tyler said, such as the overboost and underboost codes that tuners typically disable. I think some of the axes are modified too, like the optimum torque monitoring table (ECM 12414) but strangely not the main optimum torque table ECM 32900. I didn't think ECM 12414 was useful for anything other than preventing torque intervention?

    I saw a P2015 code. When you swapped intake manifolds did you remove the intake runner flaps? If so, you'll probably want to disable those codes. There are also some configuration codewords not currently available in HP Tuners that allow you to fully disable the runner flap functionality, maybe the tune you have already has those disabled if you're not seeing a check engine light for the flaps.
    Yes, regarding the P2015, the potentiometers in the different manifold is not giving the same resistance. To my knowledge the B8 potentiometer goes from 0.45 to 4.5V and the B9 goes from 1.35 to 4.15V. I did try to prototype a op-amp scaling circuit which I mounted in-line with the sensor potentiometer but I think I missed something with how the pull-up 5V is affecting the circuit. Right now I'm considering routing the wires to my arduino and scale the voltage with that, that would however need a low pass filter to try to mimic a truly analog signal from the PWM output from the arduino (if I'm not mistaken). But on the other hand it seems easier to just turn off the MIL. I wonder how this effects which maps are used though...

  20. #20
    Tuner
    Join Date
    Dec 2019
    Location
    Alingsas, Sweden
    Posts
    66
    so a bit panicky.... Took the leap to do my first ever flash with HPtuners. Just prior to flashing I updated to the latest beta version 4.7.217. The flash went according to plan, took about 3.5min to finish. Cycled the ignition and then started the car. It starts and runs but is idling a bit rough. Not to worried about that just now but almost all my channels in the VCM Scanner is gone so now I can't do a proper log!

    What do you think is this due to the new calibration or the updated VCM suit?

    EDIT: Just rolled back to 4.7.173 and all PIDs are back. Time for a test drive
    Attached Files Attached Files
    Last edited by AlexOlen; 04-29-2020 at 06:10 AM.