N52 head swap, injection and vanos controls

[Reaper0995]

If/when someone does a N52 head swap, would there be any relatively simple options to convert the injection from DI and the vanos system? Are there any other systems that would be of concern? My understanding is the DI signal isn't the same type of signal that a traditional injector requires (like those needed for the N52 head). And the Vanos on the N52 has electronic controls. Perhaps this could be disabled? I know it's ideal to use it, but the effort/cost to reward may not be worth it for a conversion like this, and hopefully the head flow would exceed any compromise from having to give this up.

 

[doublespaces]

Is this a thing?

 

[Reaper0995]

Is this a thing?

Not yet... but it looks like it is possible.

https://bmw.spoolstreet.com/threads/n54-vs-n52-block-differences.1003/

The n54 head has 31.44mm intake valves, and 22mm ex valves. Vs the N52 has 34.2 intake and 29mm ex valves. See below for visual reference.

N52:

N54:

 

[Reaper0995]

Also worth noting that the N52 has a metal valve cover.

See below for the vanos differences.

Electric actuator N52:

N54:

 

[doublespaces]

Bump. I'd love to see an n52 head on a n54 block with the n55 mevd17.

@LamboLover

 

[LamboLover]

How good is the N55 head? Fitting that with MEVD17 to N54 is more achievable DME wise I think.

 

[DennisKing]

Why?
Valvetronic would require DME control, harness modification and for what?
DI control is built into the DME and would have to be deleted
How about bolting up the turbo manifolds? Do they fit the same ports?

Would this all be for bigger valves and Valvetronic?

 

[doublespaces]

How good is the N55 head? Fitting that with MEVD17 to N54 is more achievable DME wise I think.

The N52 is port injection, so you get away from the expensive piezo injectors and direct injection limitations. The head also flows better and has a metal valve cover.

I do not care about the little advantages you get from direct injection. It has proven to cause far more restrictions than its worth.

 

[DennisKing]

If you want to run PI and little to no DI, it can be done with the N54 as it sits without swapping heads. This has already been done with the 2.0TFSI VW Engine. Your flash mapping is lower duty cycle on the DI injectors and you use a Haltech or like PI controller to run your port injectors. Basically, you make PI your primary fueling with DI backing it up. This way you don’t have to worry about the hpfp and it’s flow limits.

 

[doublespaces]

If you want to run PI and little to no DI, it can be done with the N54 as it sits without swapping heads. This has already been done with the 2.0TFSI VW Engine. Your flash mapping is lower duty cycle on the DI injectors and you use a Haltech or like PI controller to run your port injectors. Basically, you make PI your primary fueling with DI backing it up. This way you don’t have to worry about the hpfp and it’s flow limits.

The N52 head flows a lot more, it doesn't have a DI injector in the way. The N54 head kinda sucks, and from a pure power perspective so does our DI system. The idea is to do a proper DI delete more or less, better flowing head. The second part of this, is the superiority of the MEVD17. The MSD81 is nice but the MEVD17 has more potential from everything I've read.

This is really two projects in one. Using the N52 head on N54 block with MSD81 is interesting. Same with using N55 head with MEVD17 on N54 block. I'd like to have both discussions actually.

 

[The Convert]

If you want to run PI and little to no DI, it can be done with the N54 as it sits without swapping heads. This has already been done with the 2.0TFSI VW Engine. Your flash mapping is lower duty cycle on the DI injectors and you use a Haltech or like PI controller to run your port injectors. Basically, you make PI your primary fueling with DI backing it up. This way you don’t have to worry about the hpfp and it’s flow limits.

This is another band aid with an external controller lagging behind the rest of the system. What is being discussed would be a legit "fix".

 

[LamboLover]

The thing holding me back from doing a PI solution, apart from needing to finish other things first, is indeed that lag. In the idea I have for fully sequential MEVD DME controlled PI, the lag I am concerned about is the fuel that remains behind the intake valves when DI shuts off unexpectedly. With the DME controlling things, the worst case would be a single but complete port injection event has occurred before the direct injection is cancelled at the last chance. Mitigation is zero dwell, but somehow I'm not yet convinced it is the best solution even if others are happy for many such events to occur. I stop over-thinking when I'm happy with a solution - I'm not yet.

 

[The Convert]

Has anyone looked into how Toyota controls their DI PI combo? Seems to me that integrating their solution would be a step in the right direction.

 

[LamboLover]

I suspect working out the timing constraints would be very difficult because of the lack of information on Japanese ECUs. Also working out timing constraints is much tougher disassembly than almost anything else due to semi autonomous timer modules, tasks, interrupts. If we did decide that we needed to schedule DI up to 540 crank degrees earlier than now, it is a vast task to implement and test.

 

[DennisKing]

This is another band aid with an external controller lagging behind the rest of the system. What is being discussed would be a legit "fix".

I don't think that this is a Band-Aid at all. Using an actual controller (read not an AIC SS controller) would allow for this.
I think that you could probably run a Syvecs controller and handle this for less expense than swapping over an N52 head, adding valvetronic, adding the correct harness, changing your DME, and re-coding the entire car. And before you ask, yes, I do know how much the Syvecs unit is.

 

[doublespaces]

I don't think that this is a Band-Aid at all. Using an actual controller (read not an AIC SS controller) would allow for this.
I think that you could probably run a Syvecs controller and handle this for less expense than swapping over an N52 head, adding valvetronic, adding the correct harness, changing your DME, and re-coding the entire car. And before you ask, yes, I do know how much the Syvecs unit is.

And you'll still have the poor flowing head with Piezo injectors that now cost like 2 grand or something.

 

[ShocknAwe]

And you'll still have the poor flowing head with Piezo injectors that now cost like 2 grand or something.

are the DI injectors 2k for a set now? That's almost 300% of what they were... Disappointing.

 

[DennisKing]

And you'll still have the poor flowing head with Piezo injectors that now cost like 2 grand or something.

so port the head if you want more flow.

I don't think anyone is picking up on what I'm putting down here.
Run PI with 1000cc Bosch style injectors as your main form of fuel.
Run your DI as the lower volume supplement much like we do now with PI supplementing DI.
Basically backward of what anyone is running.

 

[N54QC]

so port the head if you want more flow.

I don't think anyone is picking up on what I'm putting down here.
Run PI with 1000cc Bosch style injectors as your main form of fuel.
Run your DI as the lower volume supplement much like we do now with PI supplementing DI.
Basically backward of what anyone is running.

I got what you were saying but unfortunately the big thing would be the control of the system. Most people aren't willing to spend the time and money to incorporate a separate system like a syvecs or haltech. If there was a way to incorporate the control into the DME like flex fuel, that would be ideal IMO.

 

[The Convert]

I would love to run a syvecs DME, but I'll be damned if I go down that road before a handful of other people work the bugs out.