Close

Page 1 of 8 123 ... LastLast
Results 1 to 25 of 197
  1. #1
    Join Date
    Nov 2011
    Posts
    52
    Rep Points
    132.2
    Mentioned
    13 Post(s)
    Rep Power
    2


    7 out of 7 members liked this post. Yes Reputation No

    How to increase flow rate for the N54 direct fuel injection

    Hi guys,

    The new piezo technique of direct fuel injection for the N54B30 engine has raised questions on how to increase the flow and where the limit lies. This short post tries to address (positively!!) both questions.

    I made a call to Contintental in Germany, and got to talk to one of the chief technicians of the direct fuel injection for this engine.

    This is the data that he provided me with, off the top of his head (it has been several years). He also said it would be okay for me to call him back for some more questions, but i don't want to overload him right now.

    At full load, the fuel rail has about 90 bar for the N54B30. They have tested the injectors themselves for up to 200 bar; after that point they became unstable with a possibility of leakage, something we don't want on a hot engine.

    I confronted him with our common problem of getting more fuel into the cylinder. He said the answer is NOT in raising the pressure of the fuel rail, although this would work theoretically, but due to the risk of leakage he says it's much much safer to simply change the "TI" injection time/duration. He recalled that the engine spraid about 40 mg/s (EDIT perhaps he said 14 mg/s , for mass 14.7 that would be about 27.1 lbs/m which is enough for 300 BHP ) on full load and there was PLENTY of room to play with. He recalled the duration was 0.5 - 2 ms for a spray and could safely be further increased.

    He also said that of course one should make sure the cylinder wall is not "wetted" as a result of the long spray, but he recalled surely there is a 50% margin possible for the fuel duration length.


    So the answer to our starving engine is TI. Just software ! No need to replace the piezo injectors. Piezo injectors can be embrassed by all tuners as a very flexible system with lots of room for improvement, without changing hardware.


    I would love to hear who figured out on the MSD81 how to change the TI map ? or is this already known ?

    I hope this helps and again, the info comes from a chief technician developing the N43B30 fuel injection.

    Greetings,

    Jeroen

    (Also posted on N54tech.com)
    Last edited by dubversion; 11-30-2011 at 05:08 PM.

  2. #2
    Join Date
    Mar 2010
    Location
    Toronto, Canada
    Posts
    6,683
    Rep Points
    3,335.6
    Mentioned
    225 Post(s)
    Rep Power
    34


    Yes Reputation No
    Cool info and in line with what I also thought in terms of fuel system limitations on the n54, in other words, with flashed fuel/load targets we still have no idea where the real hardware fuel limits are..
    Click here to enlarge

  3. #3
    Join Date
    Nov 2011
    Posts
    52
    Rep Points
    132.2
    Mentioned
    13 Post(s)
    Rep Power
    2



    Yes Reputation No
    I think there are really only two variables to the fuel injection system of the N54. Fuel bar pressure, set according to the tech guy at 90 bar and the duration of the injection under that pressure. Right now, it pushes out 40 mg/s under full load (can someone verify this with some maths ? )

    The pressure could be increased - as far as the injector itself is concerned to roughly double the amount currently used. However, this could be dangerous, because the chance of leakage on the overall system is increased as well; let's remember that this setup has not been tested !

    So the only other variable is the TI and apparently there's plenty to play with for an engine doing 7000 rpms !
    The piezo injector can handle it no problem whatsoever.

    greetings,
    Jeroen

  4. #4
    Join Date
    Apr 2010
    Posts
    4,616
    Rep Points
    3,236.6
    Mentioned
    24 Post(s)
    Rep Power
    33


    Yes Reputation No
    This is great to know. Thanks for the taking the time to research this and contact the right people.
    Some people live long, meaningful lives.

    Other people eat shit and die.

    I'm not racist, I hate everybody equally; especially fat people.


    Click here to enlarge

  5. #5
    Join Date
    Mar 2010
    Location
    Toronto, Canada
    Posts
    6,683
    Rep Points
    3,335.6
    Mentioned
    225 Post(s)
    Rep Power
    34


    Yes Reputation No
    If indeed correct this is awesome info indeed..

    dubversion, is the 40mg/sec a stock peak load fuel mass injected or?
    Click here to enlarge

  6. #6
    Join Date
    Nov 2011
    Posts
    52
    Rep Points
    132.2
    Mentioned
    13 Post(s)
    Rep Power
    2



    Yes Reputation No
    Yes, stock full load. This is what he recalled off the top of his head.
    What i'll do is compile all questions coming from this post (and on n54tech.com) and i'll call him up again in a few days/weeks with a follow up.
    It was a very positive talk i had with him, a true engineer !

    I'm doing this research, because i'm planning on buying the 335xi touring next year and getting it with water/meth to 500 whp. I understand that the fuel injection is one of the biggest hurdles.
    But that's off topic, let's stick to the DFI here.

  7. #7
    Join Date
    Oct 2011
    Posts
    1,058
    Rep Points
    1,149.6
    Mentioned
    39 Post(s)
    Rep Power
    12


    Yes Reputation No
    ...And everyone on e90post barked at me for saying that increasing fuel pressure is the incorrect method of increasing fuel. So what was his method of increasing either injection time or duty cycle? Depending on the limits of pulse width sent to the injectors, that may restrict the DC variability itself.

  8. #8
    Join Date
    Feb 2010
    Location
    Miami
    Posts
    286
    Rep Points
    349.4
    Mentioned
    9 Post(s)
    Rep Power
    4


    Yes Reputation No
    This is indeed good to know!
    || DCT | ESS Tune | ACM Test-pipes | AA Green Filter | Swift Spec-R Springs | Stoptech ST-40 F || || My Youtube Channel|| My Vimeo Channel ||

  9. #9
    Join Date
    Nov 2011
    Posts
    52
    Rep Points
    132.2
    Mentioned
    13 Post(s)
    Rep Power
    2



    1 out of 1 members liked this post. Yes Reputation No
    He specifically mentioned the TI, the time of injection, so a longer spray. He figured the limit is the time the intake valve opens until the time it closes, but he was merely expressing theory. But he said there should be plenty of margin to extend the TI.

  10. #10
    Join Date
    Apr 2010
    Location
    espaņa
    Posts
    749
    Rep Points
    819.7
    Mentioned
    14 Post(s)
    Rep Power
    9


    Yes Reputation No
    Click here to enlarge

  11. #11
    Join Date
    Jul 2010
    Location
    Stockholm, Sweden
    Posts
    1,104
    Rep Points
    1,398.3
    Mentioned
    16 Post(s)
    Rep Power
    14


    Yes Reputation No
    Click here to enlarge Originally Posted by m54b25 Click here to enlarge
    Also, note the piezo injector needle does & will do MULTIPLE injection per combustion =) So try to figure that one out =)
    Isnt the multi phase injection a "save the planet" feature from bmw? If one would tune/reprogram for power instead, wouldnt one rather just try to spray more ie longer?
    Click here to enlarge
    997.1 tt
    Kline 200cell exhaust
    997.2/GT2RS IC's
    Cobb E85 custom stage3 tune by Mitch
    ID1000 injectors
    Sachs stage 2.5 clutch

  12. #12
    Join Date
    Mar 2010
    Location
    Toronto, Canada
    Posts
    6,683
    Rep Points
    3,335.6
    Mentioned
    225 Post(s)
    Rep Power
    34


    Yes Reputation No
    Click here to enlarge Originally Posted by xbox_fan Click here to enlarge
    Isnt the multi phase injection a "save the planet" feature from bmw? If one would tune/reprogram for power instead, wouldnt one rather just try to spray more ie longer?
    AFAIK, and i'm obviously not a pro tuner or any sort of tuner for that matter, injection on time varies with load and RPM...the higher the RPM the shorter the time you have available to inject certain mass of fuel to satisfy the AFR target...i think that you can't have too long of an injection time at any point...if you held injectors open too long at any point you'd flood the cylinders with fuel obviously and causing misfires...
    Click here to enlarge

  13. #13
    Join Date
    Nov 2011
    Posts
    52
    Rep Points
    132.2
    Mentioned
    13 Post(s)
    Rep Power
    2



    Yes Reputation No
    dzenno, i discussed this with him. But the 0.5 - 2ms, which the engine currently uses (2ms for full load) had lots of room to spare. Imagine that the engine works at 7000 rpm, so that's 120 rounds per second. In a four stroke engine, ignition is only half the time of the rpm, so only 60 ignitions per second at full speed. Suppose that the time to spray is 1/2 of a cycle, that means that there is 1/2 * 1/60 ms time per cylinder to inject, which is 8.3 miliseconds. AMPLE room to extend the TI from the current 2 ms. Safely !

    40 mg /s spray means, at 14.7:1 means 40 * 14.7 mg / s of airflow which is 588 grams per second. This is 35.2 kg per minute, which converts to about 83 lbs of air per minute. If the rule of thumb is roughly 1 lbs for 10 BHP, i count enough fuel to make 800+ BHP !

    Perhaps his memory was incorrect and the 40mg or this mixture is bloody rich !! either way, it seems that we have ample fuel coming from the injectors.

  14. #14
    Join Date
    Mar 2010
    Location
    Toronto, Canada
    Posts
    6,683
    Rep Points
    3,335.6
    Mentioned
    225 Post(s)
    Rep Power
    34


    Yes Reputation No
    dubversion, you're spot on IMO and if 40mg/sec is indeed true those numbers show great potential in the stock fuel system..good stuff buddy
    Click here to enlarge

  15. #15
    Join Date
    Nov 2010
    Location
    Netherlands
    Posts
    6,236
    Rep Points
    6,789.3
    Mentioned
    73 Post(s)
    Rep Power
    68


    Yes Reputation No
    This post can go right into the new Tech section Click here to enlarge
    Calculating this further, 40 mg per cylinder per ignition stroke would mean the engine consumes 2.4 liters of fuel per minute at 7000 rpm.
    Isn't that a tat rich? Not sure if that's correct.
    There are two theories to arguing with women. Neither one works

  16. #16
    Join Date
    Oct 2011
    Posts
    1,058
    Rep Points
    1,149.6
    Mentioned
    39 Post(s)
    Rep Power
    12


    Yes Reputation No
    That would be a full tilt 40mg, assuming that is 800hp, does not seem unfeasable.

  17. #17
    Join Date
    Nov 2011
    Posts
    52
    Rep Points
    132.2
    Mentioned
    13 Post(s)
    Rep Power
    2



    Yes Reputation No
    I'll ask him those numbers again, about the 40 mg/s on full load next time i call him. It was off the top of his head. My calculation also shows it's quite rich (more than double what is needed for 300 BHP). But perhaps he meant the max load of the injector possible (and safe at 90 bar).

  18. #18
    Join Date
    Nov 2010
    Posts
    937
    Rep Points
    562.7
    Mentioned
    52 Post(s)
    Rep Power
    6


    1 out of 1 members liked this post. Yes Reputation No
    I think the injection window would be smaller. You calculated 1/2 cycle which would mean 360 crank deg, but with DI our window is during intake only I believe... could be smaller, or larger if you can continue injection into compression. If you calculate 180deg then it would be 4.2ms

  19. #19
    Join Date
    Feb 2011
    Location
    Indianapolis
    Posts
    95
    Rep Points
    80.1
    Mentioned
    5 Post(s)
    Rep Power
    0


    Yes Reputation No
    I would imagine that in a DI motor, our injection window is during the compression stroke not the intake stroke.
    2008 BSM 335i coupe - JB4, BMS intake, meth : Weekend warrior.

  20. #20
    Join Date
    Nov 2011
    Posts
    52
    Rep Points
    132.2
    Mentioned
    13 Post(s)
    Rep Power
    2



    Yes Reputation No
    josh, 4.2 is still double the 2.0 ms, so there's plenty of room.

    I made an edit to my original post about the 40 mg/s. I must have misheared, he probably said 14 mg/s, because for mass 14.7 that would be about 14*14.7*60/0.452 = 27.1 lbs/m which is enough for 300 BHP )
    60 = seconds to minutes
    1/0.452 is kg to lbs
    1 lbs should be good for 10.5 BHP. Considering this is DI, that number may be even higher.

  21. #21
    Join Date
    Nov 2010
    Posts
    937
    Rep Points
    562.7
    Mentioned
    52 Post(s)
    Rep Power
    6


    Yes Reputation No
    Click here to enlarge Originally Posted by indy99gpgt Click here to enlarge
    I would imagine that in a DI motor, our injection window is during the compression stroke not the intake stroke.
    The only time injection is during compression stroke is during stratified charge, which is very lean, light load. US n54s do not have this mode supposedly. under load injection is during intake.

    I think Cobb offers IPW as a channel. They also have fuel mass I believe.

  22. #22
    Join Date
    Mar 2010
    Location
    AZ
    Posts
    2,730
    Rep Points
    2,499.1
    Mentioned
    40 Post(s)
    Rep Power
    25


    1 out of 1 members liked this post. Yes Reputation No
    Good thread. Lets create some good feedback. Better yet, get the engineer's paypal. lets send him some money and let him do some digging for information we need! Seriously. Getting a chief engineer in our pockets would be game on.

  23. #23
    Join Date
    Apr 2010
    Location
    NYC
    Posts
    3,089
    Rep Points
    999.1
    Mentioned
    104 Post(s)
    Rep Power
    10


    Yes Reputation No
    Click here to enlarge Originally Posted by JoshBoody Click here to enlarge
    The only time injection is during compression stroke is during stratified charge, which is very lean, light load. US n54s do not have this mode supposedly. under load injection is during intake.

    I think Cobb offers IPW as a channel. They also have fuel mass I believe.
    Yes it does! I need to get around to logging those channels to see what it looks like. Only way I ever learn what half these logging channels mean is to actually log them.

  24. #24
    Join Date
    Nov 2011
    Posts
    52
    Rep Points
    132.2
    Mentioned
    13 Post(s)
    Rep Power
    2



    Yes Reputation No
    Click here to enlarge Originally Posted by JoshBoody Click here to enlarge
    The only time injection is during compression stroke is during stratified charge, which is very lean, light load. US n54s do not have this mode supposedly. under load injection is during intake.

    I think Cobb offers IPW as a channel. They also have fuel mass I believe.
    Hi Josh,

    Being a layman to these terms, i remember that he mentioned this. I think he said the N54 indeed uses homogeneous charge combustion, but didn't write down which type the engine uses...
    I'll ask him again next time, but i do remember that he wasn't sure either, but he was pretty sure it has to be that mode.


    About the paypal: next time i call him i'll tell him how much we appreciate his input and i'll ask if he's willing to help out some more in one shape or the other. I don't want to offend him for the kindness he showed, perhaps he's an enthousiast himself. I'll definately offer him a ride in my finished car (remember he lives in Germany with the Autobahn around the corner Click here to enlarge )

  25. #25
    Join Date
    Nov 2011
    Posts
    52
    Rep Points
    132.2
    Mentioned
    13 Post(s)
    Rep Power
    2



    Yes Reputation No
    Note to myself: IPW = ignition pulse width (= ti)

Page 1 of 8 123 ... LastLast

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may post replies
  • You may not post attachments
  • You may not edit your posts
  •