Close

    • The answer to direct injection fueling limitations/issues may be... combining port injection? KS Tuned/Kaizenspeed BMW N54 port injection manifold

      Those familiar with BMW N54 six-cylinder tuning know that the platform has been held back from achieving its maximum potential by the fueling system. Many turbocharger upgrades from dual to single solutions are available but if you can't get enough fuel into the motor what is the point of having a big turbo? The HPFP (high pressure fuel pump) for the direct injection system has a reputation of providing fueling headaches (or failing altogether) even at modest power levels.



      With direct injection the high pressure needed to inject large amounts of fuel in a short window (that gets smaller as revs go up) requires some seriously powerful fuel pumps. Thus far in the time the N54 motor has been the market (which is approaching a decade) no tuner has delivered a fully functional direct injection HPFP or LPFP solution to support the largest turbo options.

      It isn't easy (and tuners are working on making it happen with the direct injection system) but some are going outside the direct injection fuel system altogether. That is what you see here from KS Tuned / Kaizenspeed. Their solution is a port injection manifold which supplies extra fuel. Some may call this a band-aid or not a proper solution but if it works it certainly is much better than no solution at all.

      Keep in mind there are cars on the market that use both port and direct injection successfully. The most obvious example is the Toyota 86 platform which is produced by Subaru as the BRZ model and Toyota as the Scion FR-S. This chassis uses a Subaru designed 2.0 liter four-cylinder motor known as the FA20.

      Direct injection allows a fairly high 12.5:1 compression ratio thanks to the cylinder cooling effects as well as a leaner tune for fuel efficiency. Working in conjunction with the direct injection system four 205 cc Denso port injectors that supply additional fuel. Essentially a balance between a fuel efficiency design and enough fuel to make power is struck. This same sort of solution may be the answer for the N54.

      The problem though is what will control the additional injectors and coordinate with the direct injection system on the N54? Pro-Efi announced over a year ago they were working on this yet thus far no positive results have been shown. CP-E (Custom Performance Engineering) is also working on a port injection solution. Additionally, Defiv is also working on hardware for a port injection system. The problem wont' be the hardware here as much as the tuning to get a system never designed with a dual injection fuel system in mind to function.

      Does KS Tuned have the answer? We will have to wait and see. Making a computer model is easy, applying a design in real life is something else entirely. At lest we know more people are working on the N54 fuel system and eventually a solution will be available. The main question being how long is eventually.







      This article was originally published in forum thread: Aftermarket Intake Manifold started by lulz_m3 View original post
      Comments 64 Comments
      1. Sticky's Avatar
        Sticky -
        Click here to enlarge Originally Posted by uniter Click here to enlarge
        Port injected gasoline systems won't require sophisticated tuning or expensive drivers to work.
        It's a question of getting it all to play with the factory electronics and not as much as something that can control port injection. It can be done but it hasn't been done. It sure doesn't seem easy.
      1. lulz_m3's Avatar
        lulz_m3 -
        Click here to enlarge Originally Posted by uniter Click here to enlarge
        Port injected gasoline systems won't require sophisticated tuning or expensive drivers to work.

        Basically the N54 ECU is going to have a mapped load/RPM AFR target offset, supplemental injectors will fire proportional to DI duty. The ECU's built in AFR targeting PID control will take over and hit AFR to target just as quickly as with standalone DI. Controllers like the HFS-4 already measure DI as the basis for it's supplemental methanol fueling. Other controllers used by DI platforms for supplemental port injection work the same way.

        The tuning will come in to basically determine two things:
        Based on boost and injector DC - when to start supplying supplemental fuel.
        The relationship to DI Injector duty cycle.

        That is, at 500HP our stock DI fueling system can adequately provide fuel; we want the DI system to remain the primary source of fueling until additional fuel is required. We then might choose to ramp injectors up to 100% when DI rail pressure is nearing its limits, while ensuring that when supplemental injectors are at 100% that the DI system still has some ceiling. Therefore we strike on a solution where the DI system and the supplemental injectors can both be driven by the existing AFR targeting.
        The DI strategy is more complex than just DI duty cycle. It also varies the fuel pressure dependent upon load, which needs to be factored into the algorithm used to supplement the fuel. In order for the PI reference to be accurate, you would need to monitor DI DC, Dwell time/IPW, and high pressure fuel line pressure. If you ever monitor Inj DC, you will notice that it basically stays right around 60 most of the time IIRC; this is under WOT throughout the entire RPM range. I monitor with the Cobb AP, so either Cobb isn't actually monitoring the DC, or the channel is incorrectly named, or the DME is keeping DC the same but varying PW to meet fuel demands...

        Its fairly complex.
      1. rader1's Avatar
        rader1 -
        IMHO the best way to manage a PI setup is to just spray a volume that leaves the DI system plenty of room to handle precise fuel metering. Basically works like a meth kit.
      1. V8Bait's Avatar
        V8Bait -
        I'd use two banks in semi sequential (1-3-5 & 2-4-6) and start with a strategy to use them continuously based on rpm vs boost. I think if you supplement approximately 30% of the fueling, at least down low, all things would be good. Piggyback trickery would probably be helpful to keep things in line but with a clean enough PI "tune" I don't think is entirely necessary.

        I think the intake is cool but I doubt it will get 100 supporters without more info. Lol
      1. Sered's Avatar
        Sered -
        Not a bandaid at all considering Toyota uses both PI and DI on their DI-based vehicles.
      1. Sticky's Avatar
        Sticky -
        Click here to enlarge Originally Posted by Sered Click here to enlarge
        Not a bandaid at all considering Toyota uses both PI and DI on their DI-based vehicles.
        They designed it that way to begin with though.
      1. uniter's Avatar
        uniter -
        Click here to enlarge Originally Posted by Sticky Click here to enlarge
        They designed it that way to begin with though.
        TROLLING
        Running an oil catch can is a bandaid?
        Going from dual turbo to single twin scroll is a bandaid?
        Switching to colder plugs in a bandaid?
        Running more ignition advance is a bandaid?
        Replacing the clutch is a bandaid?
        Replacing the axles is a bandaid?
        /TROLLING

        On a serious note, what's wrong with bandaids?
      1. VargasTurboTech's Avatar
        VargasTurboTech -
        Click here to enlarge Originally Posted by uniter Click here to enlarge
        TROLLING
        Running an oil catch can is a bandaid?
        Going from dual turbo to single twin scroll is a bandaid?
        Switching to colder plugs in a bandaid?
        Running more ignition advance is a bandaid?
        Replacing the clutch is a bandaid?
        Replacing the axles is a bandaid?
        /TROLLING

        On a serious note, what's wrong with bandaids?
        Well basically band-aids do not fix anything, they just cover it up. A band aid approach is a term used for something that will work, but its not ideal. Its like putting a band-aid on a cut that needs stitches. Hence the term band-aid approach. PI is not superior to DI so covering up the fueling problem with PI instead of fixing the problem with a DI solution is the perfect definition of a band-aid. But hey band-aids sure come in handy when you are bleeding all over the place, which is basically the current state of our HPFP fueling problems
      1. mjmarovi's Avatar
        mjmarovi -
        Click here to enlarge Originally Posted by VargasTurboTech Click here to enlarge
        Well basically band-aids do not fix anything, they just cover it up. A band aid approach is a term used for something that will work, but its not ideal. Its like putting a band-aid on a cut that needs stitches. Hence the term band-aid approach. PI is not superior to DI so covering up the fueling problem with PI instead of fixing the problem with a DI solution is the perfect definition of a band-aid. But hey band-aids sure come in handy when you are bleeding all over the place, which is basically the current state of our HPFP fueling problems
        Yeah, but at least with PI by intake valves will stay clean!
      1. Sticky's Avatar
        Sticky -
        Click here to enlarge Originally Posted by uniter Click here to enlarge

        On a serious note, what's wrong with bandaids?
        Nothing at all here you are:

        Click here to enlarge
      1. klipseracer's Avatar
        klipseracer -
        I'm down for the Defiv plate as well. And if we can't get one person to do that then this manifold will never get support no matter how cool it is. I just want people to TRY port injection. Once there are people running around with it, it will go main stream even with an hpfp upgrade. The will compliment each other.
      1. Flinchy's Avatar
        Flinchy -
        Sad that they didn't fully remove the cyl 1 bump

        looks like it's IMPROVED, but my thinking is if they raised the runner a bit.. yeah THAT would restrict flow in itself a tiny TINY bit, but nowhere near as much as having a huge bloody dent in the side

        fix the dent and i'm in.. otherwise it'd have to be well priced :/
      1. benzy89's Avatar
        benzy89 -
        Click here to enlarge Originally Posted by Flinchy Click here to enlarge
        Sad that they didn't fully remove the cyl 1 bump
        Think I remember that the Cylinder 1 bump is stuck there because of the oil filter housing assembly

        Click here to enlarge
      1. klipseracer's Avatar
        klipseracer -
        Matt Morkin went over this a few times. You can reduce it, but the physical placement of the oil filter stuff is in the way and would require relocation which is not feasible. Maybe someone can think outside the box and get this done.
      1. Flinchy's Avatar
        Flinchy -
        Click here to enlarge Originally Posted by benzy89 Click here to enlarge
        Think I remember that the Cylinder 1 bump is stuck there because of the oil filter housing assembly

        http://www.bimmerboost.com/images/im...d3ded5c6-1.jpg
        yep... so raise cyl 1's runner and have it come in at a steeper angle?

        Click here to enlarge Originally Posted by klipseracer Click here to enlarge
        Matt Morkin went over this a few times. You can reduce it, but the physical placement of the oil filter stuff is in the way and would require relocation which is not feasible. Maybe someone can think outside the box and get this done.
        as above, why not raise the runner?
        Have it come more out of the top-ish part of the manifold and go OVER the housing.

        could hte oil filter housing be filed down slightly where it interferes?
      1. benzy89's Avatar
        benzy89 -
        Click here to enlarge Originally Posted by Flinchy Click here to enlarge
        yep... so raise cyl 1's runner and have it come in at a steeper angle?
        You really think if it was that easy, it wouldn't have already been the solution? Chances are lengthening the runners means making a bunch of other changes (like keeping all them the same length/entry angle) and just makes bigger problems in the long run.
      1. brusk's Avatar
        brusk -
        I would hope if somebody is going to go through the trouble of making an intake manifold they would make flow the primary reason and extra fuel ports a secondary benefit. Would it be possible to machine an adapter to space the housing out and forward a little to make room?
      1. SCGT's Avatar
        SCGT -
        Click here to enlarge Originally Posted by benzy89 Click here to enlarge
        You really think if it was that easy, it wouldn't have already been the solution? Chances are lengthening the runners means making a bunch of other changes (like keeping all them the same length/entry angle) and just makes bigger problems in the long run.
        Not sure what you do for work, but you'd be surprised at some of the easy solutions that are not utilized for one reason or another. Sometimes it's because it doesn't work. Sometimes it's because a guy said we need 4" clearance there when he was thinking about a hot engine component and misunderstood by some other dude working in a different section and 7 years later we're wondering why they didn't do it differently. I've seen exactly that happen many times on turbine engines that start at about about 70M/year sales value. Sure, these N54's are overall well engineered... that said it's always a very good thing to trust but verify. Envelope constraints, real or imagined, are responsible for all sorts of wonky BS.
      1. Flinchy's Avatar
        Flinchy -
        Click here to enlarge Originally Posted by benzy89 Click here to enlarge
        You really think if it was that easy, it wouldn't have already been the solution? Chances are lengthening the runners means making a bunch of other changes (like keeping all them the same length/entry angle) and just makes bigger problems in the long run.
        Why wouldn't it be that easy?.. remove the worst flow restriction, and sort out the rest later.. no? Cylinder 1 is known for being the worst flow, due to that bump... if you don't get rid of the bump, half the reason for getting an expensive AM manifold is gone, you're JUST getting better fuelling... which sure, is needed, but when it's costing thousands why do you want half the job done?

        why do you think people have thought of it, or tried it?.. That's the worst thing to assume in ANY case.

        wouldn't you think slightly changing the angle/length negative would be vastly outweighed by the positive of not having it 10-15% odd crushed? (minimum, considering the stock one is more like.. what.. 50%?)

        or if keeping them all equal length is THAT important to the designer... raise/lengthen them all slightly..?

        i remember the last time a manifold upgrade came up it was PURELY to optimise flow, with a key aspect being removing the dent?

        Click here to enlarge Originally Posted by brusk Click here to enlarge
        I would hope if somebody is going to go through the trouble of making an intake manifold they would make flow the primary reason and extra fuel ports a secondary benefit. Would it be possible to machine an adapter to space the housing out and forward a little to make room?
        +1.

        and what @SCGT said.. it's never safe to assume someones interpreted/implemented it right.

        there HAS to be a way to fix the manifold/runner/get it all working well.. raise the whole plenum up slightly? just have them ALL arch up (which wouldn't be good for 2-6 though no?)

        or just try and raise cylinder 1's...

        if they've done all the CFD on it, and shown that this is absolutely the optimal manifold... then fair enough.
      1. leveraged sellout's Avatar
        leveraged sellout -
        This is extremely interesting, as it would be very cool to have such a solution, but if there are other solutions coming that implement the DI system....well I'm just excited to see the potential. N54/N55 stuff in general is way more exciting than the same old S63 powered car with exhaust, HREs, and tints. So anything new and interesting in this space should be interesting.

        Costs don't have to be too crazy with this so long as the controller is kept within reason. That's the hard part, really. But even so...I'd love to see what kind of numbers this could support.