Port Injection?

[AB M2C]

Need a little help.
Who here is running port injection?
Which kit are you running and why did you choose it?
What LPFP did you choose?

Big question though is what are you using to control it? are you using the JB4 add on? or the Split Sec AIC6 Port Injection Controller? Something else?


How did you tune it as well?

Any help is appreciated here.

[ShocknAwe]

I'm actually doing research on this now too. I'm not sure what you're using for tuning, but I'm a JB4/MHD backend user, so that influences my thoughts.

I was thinking a Stg2 LPFP with the BMS/FuelIT port injection kit would be key with the JB4 integration for easiest setup and tuning. BMS stopped supporting the AIC controller due to tuning issues I think.

Other option which I would feel comfortable going with is the Precision Raceworks kit, though this uses the MMP port injection rail, which has had some fairly major problems through development. I think Mauricio worked it all out, but still.

[AB M2C]

Originally Posted by ShocknAwe

I'm actually doing research on this now too. I'm not sure what you're using for tuning, but I'm a JB4/MHD backend user, so that influences my thoughts.

I was thinking a Stg2 LPFP with the BMS/FuelIT port injection kit would be key with the JB4 integration for easiest setup and tuning. BMS stopped supporting the AIC controller due to tuning issues I think.

Other option which I would feel comfortable going with is the Precision Raceworks kit, though this uses the MMP port injection rail, which has had some fairly major problems through development. I think Mauricio worked it all out, but still.

Thanks for the thoughts.

Ya right now I have a vishnu procede which is limiting. (Its great for anything FBO, but beyond that it quickly becomes obsolete. I've had it for a long time and I certainly wish I had a JB4 due to the continued support and extra versatility, but thats where I'm at). I'm considering ditching it and going for the JB4 simply because of this, and I'm not sure how easy it would be to run the AIC6 in conjunction with it.

I'm pretty sure that MMP has worked out there issues, but I'm still a little leary. Thats actually the reason I started this thread to see if anyone has had a good experience with that one, or if I should just go with Fuel It kit.

For LPFP right now I'm running the presicion racewerks modular kit with their 450, but also a "jerry rigged" 250 piggy back, so my fueling I feel is sufficient for more power than I want. Just need to get passed the stupid HPFP, and I should be good.

[ShocknAwe]

Cost for cost given the price of updating to PI, I'd probably just go ahead and order the JB4 G5 ISO and be done with it. After reading on the AIC vs JB4 controller more today I'm not sure I'd put the AIC on.

What I don't know is what's involved in hooking the MMP kit up with the JB4 controller.

[bNks334]

I've been doing research on this too. Don't touch anything less than direct port injection.

As for controllers, I have no confidence in anything offered. Every one of the various controllers is REACTIONARY meaning there will always be time (ignition cycles) before PI starts/stops spraying when DI has already started/stopped.

Only true safe way to add supplementary fueling to direct injection is to have the DME controlling it (aftermarket DME) which no one has accomplished (correct me if I am wrong on this)... The capability isn't programmed into the stock DME from what I understand. What we have now WORKS (except when it doesn't).

JB4 boasts being able to integrate into the can-bus and monitor various aspects of fueling (pressure) and performance (O2 data, boost pressure, etc..). Most of those issues BMS mentions are inherent to JB4 use only... JB4 is biasing the stock signal in the first place which prevents the DME from seeing proper boost/load and correcting them. Therefore, Terry's comments are a bit misleading in boasting meaningless features that wouldn't be all that necessary in the first place if you weren't using a JB4 LOL. Not to mention, the JB4 itself is still a reactionary system and can-bus communication is entirely too slow to rely on the data for any kind of added "safety." I call BS all around.

I myself will probably invest in HPFP upgrades when I move to a larger turbo. The expense of adding on port injection, controllers, efforts to tune, and adding massive amounts of additional risk/failure points far exceeds just running something like VTT's overdriven HPFP systems. To each there own I guess... some people just can't get away from their love for their JB4' and for whatever reason the community insists on making things work AROUND JB4 flaws instead of scrapping it. Damn those gauge hijacking features are cool...

Hopefully we have some real fueling options soon in the form of HPFP upgrades and rail uprgades. What we have now is not up to par with what people are trying to do... or with other platforms.

Best thread by far where people are actually willing to discuss the shortcomings of current PI controllers:

https://bmw.spool
street.com/threads/pi-controller-pros-cons-comparison-jb4-pi-vs-split-6.1038/

[AB M2C]

Originally Posted by bNks334

I've been doing research on this too. Don't touch anything less than direct port injection.

As for controllers, I have no confidence in anything offered. Every one of the various controllers is REACTIONARY meaning there will always be time (ignition cycles) before PI starts/stops spraying when DI has already started/stopped.

Only true safe way to add supplementary fueling to direct injection is to have the DME controlling it (aftermarket DME) which no one has accomplished (correct me if I am wrong on this)... The capability isn't programmed into the stock DME from what I understand. What we have now WORKS (except when it doesn't).

JB4 boasts being able to integrate into the can-bus and monitor various aspects of fueling (pressure) and performance (O2 data, boost pressure, etc..). Most of those issues are inherent to JB4 use only (like needing to see boost target since JB4 is biasing the stock signal in the first place)... so his comments are a bit misleading. Not to mention, the JB4 itself is still a reactionary system and can-bus communication is entirely too slow to rely on the data for any kind of added "safety." I call BS all around.

I myself will probably invest in HPFP upgrades when I move to a larger turbo. The expense of adding on port injection, controllers, efforts to tune, and adding massive amounts of additional risk/failure points far exceeds just running VTT's overdriven HPFP systems. To each there own I guess... some people just can't get away from their love for their JB4's.

I was wondering about the reactionary. Do you really think that either computer takes so much time the mixture could get lean enough to detonate?
I don't like how it is reactionary, but my issue with the VTT is (and maybe I'm wrong about this) but it still only maxes out at 550ish hp? Then you have to add another HPFP to do any more? That gets pretty expensive pretty quick, and it seems like a lot of stress on an already prone-to-failure component. Also to be honest I'm not sure where it comes into play or how its not reactionary as well but you need an external controller for the VTT if you plan on running the piggyback pump yes? So I don't know why this wouldn't have the same issues as a port injection controller. (Again I could be wrong about this so don't 100 percent take my word).

Neither option is perfect, but I was hoping that port injection programming (with integration to the JB4) had come far enough that it was plenty safe to run now. Apparently I could be wrong about that and a simple yet expensive HPFP upgrade is the way to go?

Anywho, thanks for the info definitely another path that is worth considering.

[bNks334]

Originally Posted by ThatOneKid

I was wondering about the reactionary. Do you really think that either computer takes so much time the mixture could get lean enough to detonate?
I don't like how it is reactionary, but my issue with the VTT is (and maybe I'm wrong about this) but it still only maxes out at 550ish hp? Then you have to add another HPFP to do any more? That gets pretty expensive pretty quick, and it seems like a lot of stress on an already prone-to-failure component. Also to be honest I'm not sure where it comes into play or how its not reactionary as well but you need an external controller for the VTT if you plan on running the piggyback pump yes? So I don't know why this wouldn't have the same issues as a port injection controller. (Again I could be wrong about this so don't 100 percent take my word).

Neither option is perfect, but I was hoping that port injection programming (with integration to the JB4) had come far enough that it was plenty safe to run now. Apparently I could be wrong about that and a simple yet expensive HPFP upgrade is the way to go?

Anywho, thanks for the info definitely another path that is worth considering.

Ignition cycles happen in milliseconds. No, I personally don't believe the JB4 can poll through the can-bus, process the data, and pass data back along the can-bus within a single ignition cycle. Not happening.

You are right, you quoted the limit of a single over-driven HPFP. VTT has a double HPFP system as well which just made 800whp+ on their stage 2 turbo upgrade. Can't speak for the specifics on how the double barrel shotgun system works... I was more stating that HPFP upgrades is the direction the community should be pushing... not slapping on "good enough" port injection without any real correlating control over direct injection... I was not promoting VTT and I have no experience with their product.

The issue with a HPFP being driven by a controller is NOT the same as port injection. Port injection can continue to spray when the DME has shut down individual cylinders (something the N54/N55 dme is capable of doing as a part of normal operating conditions for performance, efficiency, and safety). Yes, there will be a delay even with JB4 "fail-safes."

Secondary HPFP control just means that rail pressure may see some spikes/dips if something goes wrong or gets delayed. DME maintains its 100% control over all fuel injection (direct injection) and can safely shut down cylinders as it needs to without issue. A swing in rail pressure will not present much performance issues and can be tuned for (like building a bit pf excess pressure with the primary pump right before the rpm break-point where the secondary pump engages...).

I don't know how to say the above any more clearer than that lol

[bNks334]

This is just one persons opinion though so don't let me scare you out of PI. Plenty of people are running PI without issue. Misfires shouldn't be happening in the first place I suppose.

[ShocknAwe]

An overdriven HPFP is ABSOLUTELY another worthwhile consideration. To be honest, I think both systems have serious drawbacks right now. I think you're right in saying there's no option for port injection which is controlled directly by the DME. And then on the other hand, a HPFP which has been moved over and driven at engine speed for an increase of about 12% flow in exchange for 300% operating speed is an unknown variable. VTT has been doing it for a while and in speaking with Chris over there, he is unaware of any failures, but there aren't enough folks daily driving on shotgun HPFP setups to know for sure.

One of the big reasons I'm considering a port injection solution is to put fuel on my valves. I've seen pictures from folks who have used PI since their last walnut blast and it's almost enough to convince me alone. I need to open up my IM and check but I'm sure my valves are completely disgusting again.

Any way you cook it, if you go north of 600whp you're asking for trouble in the long run in terms of engine internals. I'm not sure a reactionary system is going to be all that big a deal at 500-600 levels as long as the end user MAINTAINS the car. If you get into talking about the guys running huge HP, there are a good many of them using semi-defeated knock detection. Want to talk about stupid? Ehhhh...

Here's another question. How much weight does a PI system add to the car vs a single-barrel shotgun pulley HPFP? If the HPFP is old and showing signs of struggle, that could be a good argument for heading down the VTT side of the solution.

[HawkeyeGeoff]

On the DISI Mazdaspeed platform we ran both 5th/6th port (basically inserting a coupler in the charge pipe and running via methanol style) and direct port injection; both of these systems are used in conjunction with a Split Second Controller. That car is MAF based so you basically just spliced it into the MAF voltage wire, spliced into the coil packs and ran a table for injector pulse width (or injector duty cycle) at whatever MAF voltage you wanted. In my case in particular, I was running a separate fuel cell. I used an analog boost based switch to trigger the separate fuel pump to start up at a low boost ~6 psi to ensure it would be primed and ready when it saw the right MAF voltage.

Since the car was MAF based (this is relevant for N55) you would just adjust the MAF cal to achieve the correct AFR and DI IDC that you wanted. This would be very easy using a Cobb ATR style system but I really have no experience using MHD. I assume it does it similar and JB4 just automatically adjusts the MAF cal based on this.

Both styles of the system will work for sure. It seems in general that BMW owners have it easy in regards to tuning....all the work is basically done for you. In the little time that I owned my N54 I was pretty stunned that you really could just piggyback and let it do its thing.

Regardless as long as the system is primed and ready, I see no real problems using the Split Second controller to do the same thing. The "delays" you are saying really aren't a thing when you're relying on injection systems when you aren't running it through a JB4.

[bNks334]

Originally Posted by HawkeyeGeoff

On the DISI Mazdaspeed platform we ran both 5th/6th port (basically inserting a coupler in the charge pipe and running via methanol style) and direct port injection; both of these systems are used in conjunction with a Split Second Controller. That car is MAF based so you basically just spliced it into the MAF voltage wire, spliced into the coil packs and ran a table for injector pulse width (or injector duty cycle) at whatever MAF voltage you wanted. In my case in particular, I was running a separate fuel cell. I used an analog boost based switch to trigger the separate fuel pump to start up at a low boost ~6 psi to ensure it would be primed and ready when it saw the right MAF voltage.

Since the car was MAF based (this is relevant for N55) you would just adjust the MAF cal to achieve the correct AFR and DI IDC that you wanted. This would be very easy using a Cobb ATR style system but I really have no experience using MHD. I assume it does it similar and JB4 just automatically adjusts the MAF cal based on this.

Both styles of the system will work for sure. It seems in general that BMW owners have it easy in regards to tuning....all the work is basically done for you. In the little time that I owned my N54 I was pretty stunned that you really could just piggyback and let it do its thing.

Regardless as long as the system is primed and ready, I see no real problems using the Split Second controller to do the same thing. The "delays" you are saying really aren't a thing when you're relying on injection systems when you aren't running it through a JB4.

You ignored the part where N54/N55 dme can shut down individual cylinders and your PI will continue to spray...

What you mention from your MS3 days is just as disconcerting... there is no reliability in injecting anything via the chargepipe... How can you ensure fuel homogeneity across all cylinders? You are also literally turning your intake into a bomb in the event that the car backfires. Rich conditions will also occur when throttles closures happen and/or DME shuts down cylinders... All bad things for reliable and consistent tuning for a daily driver and not a one time dyno queen.

I was also under the impression that MS3 suffered from injector limits before HPFP limits... the best, and more expensive, option on that platform is to upgrade to better injectors. The charge-pipe injection being more of cheap hack job... Same thing on an N54/N55... you can install a meth kit dirt cheap in the charge-pipe for a quick fueling fix for lower whp #'s. Meth has proven to not provide enough supplementary fueling to run 600whp+ though. The injector nozzle size required to inject that much meth causes wetting to occur and misfires. It's not reliable at all.

[ShocknAwe]

Can't say anything about the flow pattern or distribution of the MS3, but on the N54 spraying into the TB or CP is really not a good idea. Airflow through the IM will not equally balance fuel at all, Cylinders 5&6 are particularly bad. I'd do PI or DI only.

Now if we're talking about spraying a TINY amount of methanol into the charge air to cool it down a little, sure, shoot that into the CP directly after the FMIC and call it a day.

[The Wind Breezes]

If you have a legitimate need for port injection you're already spending quite a lot of money on parts labor and tuning. People are making more than double stock power with the stock DI.

Don't fuck it up by doing anything less than the right way. If you don't use one injector per runner you will have inconsistent fueling across your cylinders. If you use a JB4...well, if you do that you probably deserve whatever happens.

At the VERY least you will need a system that reads fueling directly from the DME with no lag. This is not that simple of a matter because not only do you have the on time for the injectors but you have rail pressure as well. A JB4 or other piggyback will not do this.

Really, if you have a *legitimate* need for port injection (and keep in mind people have made quite a shitload of power with DI) you might as well just run a standalone ECU and ditch the DI fueling. Also are you SURE you need port fueling? What kind of power goals do you have?

[bNks334]

Originally Posted by ShocknAwe

An overdriven HPFP is ABSOLUTELY another worthwhile consideration. To be honest, I think both systems have serious drawbacks right now. I think you're right in saying there's no option for port injection which is controlled directly by the DME. And then on the other hand, a HPFP which has been moved over and driven at engine speed for an increase of about 12% flow in exchange for 300% operating speed is an unknown variable. VTT has been doing it for a while and in speaking with Chris over there, he is unaware of any failures, but there aren't enough folks daily driving on shotgun HPFP setups to know for sure.

One of the big reasons I'm considering a port injection solution is to put fuel on my valves. I've seen pictures from folks who have used PI since their last walnut blast and it's almost enough to convince me alone. I need to open up my IM and check but I'm sure my valves are completely disgusting again.

Any way you cook it, if you go north of 600whp you're asking for trouble in the long run in terms of engine internals. I'm not sure a reactionary system is going to be all that big a deal at 500-600 levels as long as the end user MAINTAINS the car. If you get into talking about the guys running huge HP, there are a good many of them using semi-defeated knock detection. Want to talk about stupid? Ehhhh...

Here's another question. How much weight does a PI system add to the car vs a single-barrel shotgun pulley HPFP? If the HPFP is old and showing signs of struggle, that could be a good argument for heading down the VTT side of the solution.

IF the single barrel can support 500-600 whp on 100% E85 to 50% E85, respectively (hard to prove so far), then that is far more than a 12% increase in flow. Not sure if you were just using numbers to exemplify the extra stress on the HPFP or you were quoting actual numbers you found somewhere though lol

[HawkeyeGeoff]

Originally Posted by bNks334

You ignored the part where N54/N55 dme can shut down individual cylinders and your PI will continue to spray...

What you mention from your MS3 days is just as disconcerting... there is no reliability in injecting anything via the chargepipe... How can you ensure fuel homogeneity across all cylinders? You are also literally turning your intake into a bomb in the event that the car backfires and causing rich conditions to occur when throttles closures happen and/or DME shuts down cylinders... All bad things for reliable and consistent tuning for a daily driver and not a dyno queen.

Well....it works and has been proven to work on lots of high HP DISI cars. In my 8+ years on that platform I never saw one go boom from that.

For the 5th/6th port style you can't ensure homogeneity; hell for DIRECT port injection you cannot really expect it either due to flow differences in the intake runners. It's better but still not the best due to injector angle / matching of the injector to the flows, etc.

I'd put money on it actually working just fine when it's implemented correctly; maybe there's just a better option out there right now with the single barrel with this platform.

You should also check out your min AFR after you lift after a WOT pull up top; I bet you'd see a min lower than your target even on a stock DI setup. It's near impossible not to get a rich AFR reading after a WOT, lift and throttle closure. As long your injectors are shutting off when they're supposed to it should be fine.

[HawkeyeGeoff]

Originally Posted by ShocknAwe

Can't say anything about the flow pattern or distribution of the MS3, but on the N54 spraying into the TB or CP is really not a good idea. Airflow through the IM will not equally balance fuel at all, Cylinders 5&6 are particularly bad. I'd do PI or DI only.

Now if we're talking about spraying a TINY amount of methanol into the charge air to cool it down a little, sure, shoot that into the CP directly after the FMIC and call it a day.

Say that to the dudes running two D14's

Also. http://www.fuel-it.biz/n54-throttle-body-injection/

Guess 5th/6th port / methanol style injection is a thing on this platform as well.

[bNks334]

Originally Posted by HawkeyeGeoff

You should also check out your min AFR after you lift after a WOT pull up top; I bet you'd see a min lower than your target even on a stock DI setup. It's near impossible not to get a rich AFR reading after a WOT, lift and throttle closure. As long your injectors are shutting off when they're supposed to it should be fine.

You maintain your point while ignoring mine lol. The supplemental injectors DON'T shut off in time. Issues are more concerning when this happens at WOT, which they do on this platform, not necessarily on lift... Hydrolock much? JB4 is a cool piggy back, but the scope of this conversation is way beyond the abilities of the JB4 as a TUNER. We are talking about custom flash tuning and using JB4 as a port injection controller which it is also capable of doing. MHD is not a piggy back, it is an android app used to flash modified DME files. MHD can't scale anything. The shops who are tuning for big power, who the people on this platform pay small fees to to professionally "tune" their cars via flash tuning, openly acknowledge the drawbacks and inherent flaws of port injection without adequate DME control. I already linked to a good discussion above you probably didn't read.

The DISI platform maxes out injectors before maxing out the HPFP. There is no option but to add supplemental injection since there is no injector upgrade available... That does not mean that it's an ideal solution and that people are not ignoring the inherent risks in making it work... wouldn't you upgrade your injectors if you could instead of adding PI? People HAVE lost motors in the BMW community and, I am sure they have in the MS3 community as well. Whether people A) want to admit it ,or, B) Even know/understand the root cause... is two things to consider on both these platforms.

[bNks334]

Originally Posted by HawkeyeGeoff

Say that to the dudes running two D14's

Also. http://www.fuel-it.biz/n54-throttle-body-injection/

Guess 5th/6th port / methanol style injection is a thing on this platform as well.

Never said it wasn't, and I never said it didn't work. Data has been posted in this community that shows that meth injection at the throttle body or chargepipe does not provide for a homogeneous fuel mixture across all cylinders. As stated previously, this is due to the intake manifold design. Also, like I previously stated and you are still ignoring, meth injection alone is not enough supplemental fueling to make big power whether done through the charge air (problems also include wetting and homogeny) or direct port injection...

Try running an 800whp motor with throttle body injection. There is someone who does it on 600whp+ and he has blown up his intake manifold twice... Also, going back to the fuel mixture being homogeneous, some cylinders may see a fuel mixture of 100aki and others only seeing 95aki (as an example). Throttle body injection does not provide for reliable tuning. Period.

Again, it works.. no denying that... it doesn't mean that it's the direction the platform should be taking to achieve high WHP goals when better options may exist. I think everyone posting here knows what the available options are. No need to post up links to throttle body injection. It's not a good fueling solution for big power goals. Throttle body injection is bottom of the barrel next to injecting at the chargepipe.

[AB M2C]

So to sum up:

After going through all of this, and doing quite a bit more research: I think a boosted HPFP is actually going to be the way for me.

Advantages of port injection are the scalability and honestly a big thing (brought up by ShocknAwe) is that you get a free valve cleaning every time fuel flows through the intake valves. In my opinion thats a huge advantage. I walnut blast this car roughly every 10k, which may be slightly excessive, but every time there is a pretty substantial buildup on the valves. Its not the biggest deal in the world to me anymore because I've done it several times, but it would be nice to never have to do that again, and have piece of mind the intake side is running at 100% ish efficiency...

Drawbacks of that system: the tuning, as stated by bNks334 it is a reactionary system so there will be some lag, he even went though the electronic pathway, but with a JB4 and the JB4 port injection add on it should be tunable enough to not create catastrophic issues with leaning out and detonating. Still there will be some delay and tuning this system can be difficult and potentially dangerous if done wrong. I'm not the best tuner, but do prefer to do my own work, so this did influence my decision.

The advantages of going the boosted HPFP route are there is no secondary/extra tuning needed so it shouldn't be a reactionary system, rather it is controlled by the DME and everything stays DI. This is almost definitely the safer option for engine health and a much easier system to tune. Also (with the piggyback pump) you should be good to the 800 hp range which as I'm shooting for mid-high 600's this is plenty enough for me. Unfortunately the single barrel option will not be enough, so I'll have to go full double barrel. Another advantage from VTT is supposedly even if your HPFP fails your car will go into limp mode so you won't lean the fuel mixture, and you can limp the car until you fix the pump. Having extra safety mechanisms in place is never a bad thing in my mind.

Drawbacks: you are running a secondary pump of the motor, there will be extra parasitic power loss which is inherent to anything belt driven off the motor. Also there is extra modification required to the coolant routing as well as to the AC system, due to the addition of the secondary pump. It is difficult installation as you even have to drop a motor mount, but can be done if you have some know how.

Overall though I'm going to agree with bNks334, the advantages of upgrading the HPFP system in tunability outweigh the advantage of PI. If there were 1) a DIRECT replacement for the HPFP making easy of installation no big deal, or 2) an ECU that could control PI injectors individually, that was able to monitor the DME this would be an easy decision to go one direction or the other. As neither of those are options right now though, I think the safety in going with the overdriven HPFP is the better option than going with the extra fuel blast of PI.

Hopefully I've actually summed stuff up right. If I'm wrong anywhere let me know. hopefully this will actually be a useful thread to someone else at some point, but it definitely helped me. Thanks ShocknAwe, and bNks334, awesome information from both of you.