Direct injection and mods

With so many new cars coming out with direct injection, how tunable are these systems for the DIY crowd?

For example with a n/a to turbo application, could you stretch the stock injectors further since fuel can be injected over a much longer period?

This one is easy! The answer to your questions: "It depends..."

Bryce

I used to own a Mazdaspeed3. It's factory turbo but you might run into this issue with an N/A DI engine depending on how it is set up. The injectors had plenty of overhead if what I've been told is correct. However, the camshaft-driven, high-pressure fuel pump (HPFP) would become less effective over 5500rpm.

IIRC, the stock MS3 HPFP had a max. fuel pressure of 1800psi between 0-5500rpm, but would slowly peter out until redline, where it was more like 1500rpm. On a hard pull, the car felt like it had run out of breath over 5500rpm. The MS3 was safe when stock or with a mild Cobb Accessport tune, but if you installed a downpipe and/or FMIC, or bumped up the boost, it was highly recommended to upgrade the HPFP internals for a bigger safety margin so the engine isn't starved of fuel. It's a relatively easy job. After installing KMD brand HPFP internals, fuel pressures on my car stayed near a consistent 2000psi throughout the rev range. I didn't have to bitch shift at 5500rpm anymore.

DI Audis and VWs use a similar HPFP and have a similar issue. There are many VW/Audi tuners besides KMD that make HPFP internals and they all fit the Mazda because the pumps are the same internally. Some BMWs with the N54 (DI, twin-turbo) I-6 had major HPFP issues leading to engine failures and BMW extending warranties on the HPFP.

One other thing to watch out for is carbon buildup on the intake valves. Multi-point fuel injection washes carbon off the intake valves, but DI doesn't since it's injecting straight into the cylinder. Cleaning intake valves is no fun. IIRC, some newer designs include both MPFI and DI. The MPFI is mainly there for cleaning valves and for supplementary fueling in high load situations, while DI is on all the time.

Nashco wrote: This one is easy! The answer to your questions: "It depends..." Bryce

Exactly- what are you trying to tune?

RexSeven addresses the biggest issue with tuning DI engines, and that's the capacity of the high pressure pump- since it's volume limited, there will be a point where it will no longer be capable of delivering the fuel mass that you need.

That amount has been creeping up in recent years, but generally, they all limit out between ~350-400hp of fuel flow (you can back calculate the fuel flow from that, but it's generally there). More than that, and you need a second pump. Or a PFI fuel system to add to it.

So, for an NA DI engine, what do you want to do?

As for tuneability in general, while there are companies out there who do some tuning, what I've seen is 1) generally they are hackers and 2) it's getting more and more difficult to do, especially to retain warantee. The one person we dealt with- he just played with numbers, not really knowing the effect, and drove the 1/4 until the engine blew up....

Stumbled across this company the other day. Probably not aimed at amateur tuners, but the parts are out there. http://www.liferacing.com/?page_id=35

ncjay wrote: Stumbled across this company the other day. Probably not aimed at amateur tuners, but the parts are out there. http://www.liferacing.com/?page_id=35

Interesting- should not be too shocked that someone finally has come up with a DI injector driver. Now one needs the high pressure pump control to go with it.

(and the tuner also needs to know that injection timing is critical now- in addition, it MUST be fully sequential, and in no way batch injection.)

Has anyone noticed what the BMW guys are doing with the newer DI Turbo engines? From what I recall, the 335i would peter out at around 400whp on a reflash (if the injection pump would hold up even that long)

Hmmmmm.... The di/pi combo is very interesting. anyone know a good place to learn more (beside teh googels)?

bluej wrote: Hmmmmm.... The di/pi combo is very interesting. anyone know a good place to learn more (beside teh googels)?

Interesting, but very complex. what do you want to do?

DI/PFI still results in the rather significant complexity (for a DIY'er) of DI, and trying to map when you want to do both.

If you want REAL complexity- think PFI-gas, DI- E85- there you get some serious compelxity.

Oh, i don't doubt the complexity. I also don't have a specific application in mind, was just musing at the idea of adding the pfi in general to an oem dfi only system. Just curious how its been implemented in the past.

bluej wrote: Oh, i don't doubt the complexity. I also don't have a specific application in mind, was just musing at the idea of adding the pfi in general to an oem dfi only system. Just curious how its been implemented in the past.

Toyota did it, as they seemed to use it more for an emissions need vs. a performance need. You might start see it in the future, particularly in Europe, as the new PM standards start ramping up.

In theory, you can use a combination pfi/di on a cold start to get the best emissions, but it would take some time to determine the right timing for both set ups. For peak power, I'm not sure of the percentage of di fuel you need to get all of the knock benefit of di- but you could split it so that it does not max out the high pressure pump.

Adding pfi to an already existing di set up would require a brand new computer, since you need 4,6, or 8 new injector drivers for pfi, and an entierly separate fuel rail system for the pfi injectors. It would require a lot of dyno time to understand when to blend the percent di/pfi used all the time.

Or at least a very sophisticated piggyback computer to know when to run the PFI set up. Possible, and maybe required for some engines, but, well.... It's only money.

I guess I was thinking someone with the capability to write to the stock ecu like some of the aftermarket is doing, might be able to set the dfi to work in a consistently and desirable enough manner that you could then add something like MS with dedicated fuel pump/lines/rail/injectors.

Sure, not the simplest/cheapest but if you're after more power than the stock hpfp can supply, it may not be out of line.

Or I could be talking out my arse, but hey, those are thoughts I have.

In reply to bluej:

If you can find someone who can do that sophisticated job with the OEM computer, I would suggest that an MS would be way, way, way down on their skill level. For a hack jobber, MS would be about it, though.

The real key is to make them work together, if you can do that, it can be done.

RexSeven wrote: One other thing to watch out for is carbon buildup on the intake valves. Multi-point fuel injection washes carbon off the intake valves, but DI doesn't since it's injecting straight into the cylinder. Cleaning intake valves is no fun. IIRC, some newer designs include both MPFI and DI. The MPFI is mainly there for cleaning valves and for supplementary fueling in high load situations, while DI is on all the time.

Sorry to revive an older thread, but I have been seeing frequent reports of carbon build up issues with these turbo DI engines on the BMW message boards and the need for frequent valve cleaning treatments. I assume the build up is caused by the PCV system and wonder if an oil catch can would help alleviate this condition. I was planning to replace my Mazda3 with a Mazdaspeed3 and am trying to understand this issue better.

Also, does anybody know if the new Focus ST Ecoboost engine is setup with both DI and MPFI?

Thanks in advance, John

In reply to killeen_john:

That I am aware of, the issue is mostly a BMW problem, and not a DI problem. I've not heard of any problems with our motors, nor GM's, nor Toyota's, nor a number of other in terms of intake system deposits. Except for BMW. And they've had problems with that for at least 30 years, since I was working on it as a project at Cheveron in the late 80s.

The Focus ST motor is DI only. No good reason (at the moment) to have both.