XDI High Pressure Fuel Pump upgrade

[andrewtac]

Vortech.

I can’t answerthe question either but I will say I have my doubts as to whether most tuners are even capable of properly tuning for for the Xdi pump. After many months I’m still learning but I think I have it under precise control now. FYI I’m running 3250 psi pressure (stock is 2900) and have learned a lot about how much EOI timing can affect power.

I'll consult Lund prior to the purchase. I know they tuned a beefcake car with the pump; but that wasn't exactly a street car.

Sponsored

 

[4sdvenom]

Sorry I keep doing that, assuming everyone knows; vortech V7-JTB; finally ordered it 4-6 week wait.

Got it!
My fault for not following the centrifigals more lol.
Thanks

 

[illtal]

I raised my boost 2 psi (3.75" pulley, Whipple 3.0) and found the limit of the XDI pump/stock injector combination. But this comes with a few caveats.

I could no longer maintain 90% GDI blend in two scenarios:
1. Ambient 73 deg, COT enrichment ON, 7000 rpm, 1.65 load. On that run, GDI blend fell to 83% min.
2. Ambient 57 deg, COT enrichment OFF, 6900 rpm, 1.68 load. On that run, GDI blend fell to 80% min.

However, this really doesn't demonstrate the absolute limit of the XDI HPFP45 for a few reasons:
1. I'm still using the stock injectors and 2900 psi GDI pressure. Larger injectors would allow a lower pressure to be run, which would result in more flow available. Note that the Gen3 coyote GDI injectors are very small - much smaller than the 3.5 and 2.3 ecoboost engines.
2. I'm running .75 lambda, then .72 in COT mode. Without cats, you could run at .82 or leaner, which would allow over 10% more power to be made with the same fuel flow.
3. I'm attempting to hold 90% GDI to get the maximum detonation suppression. Lowering the command to 80 or 85% would also reduce the demands on the GDI system.

Something else in the log was that my DW400 is officially at its limit at the same time. DC was max'd at 50% and pressure fell below 60 psi, so to pushing things further would probably require a DW440 or BAP, and a set of larger GDI injectors.

@engineermike can running larger di injectors without the xdi pump be beneficial?

 

[engineermike]

@engineermike can running larger di injectors without the xdi pump be beneficial?

Definitely not. The stock injectors will flow enough that the pump is max’d out well before the injectors are a factor. I do think more is available from the stock pump if you play with the pressure, though, which is something almost nobody seems to go after.

Since my post on 4/23 I’ve learned more. After that post, I had been lowering the gdi pressure to get more pump flow, and moving EOI (end of injection) further back in the cycle to get more time for injection at the lower pressure. I had gdi flow optimized at 2600 psi and EOI occurring at 60 deg BTDC firing (about 40 deg before spark), maintaining 90% blend even at high speeds and loads. While I had found technical papers plus practical data indicating up to 20 deg before spark was fine, I learned quite the opposite. Injecting from 330 to 60 deg BTDCf was absolutely killing my top end power. I’m talking 100++ hp loss at 7500 rpm. I got all the power back by clipping EOI at 110 deg BTDCf, which is still much later than the GM crowd believes is good. They say keep all the injection in the intake stroke (>180 EOI).

Anyway, pushing EOI that much earlier really shrunk my injection window and killed the blend, so I had to increase gdi pressure to 3250 psi to get the blend back up. As it stands, I’m bottoming out at 70-75% blend which is still lower than I want. I’ve been kicking around upgrading gdi injectors so I can maintain my 110 deg EOI but get more flow in a shorter period of time, which would allow higher blend at high rpm (when time is limited) and also more pump capacity by lowering its pressure.

 

[illtal]

Definitely not. The stock injectors will flow enough that the pump is max’d out well before the injectors are a factor. I do think more is available from the stock pump if you play with the pressure, though, which is something almost nobody seems to go after.

Since my post on 4/23 I’ve learned more. After that post, I had been lowering the gdi pressure to get more pump flow, and moving EOI (end of injection) further back in the cycle to get more time for injection at the lower pressure. I had gdi flow optimized at 2600 psi and EOI occurring at 60 deg BTDC firing (about 40 deg before spark), maintaining 90% blend even at high speeds and loads. While I had found technical papers plus practical data indicating up to 20 deg before spark was fine, I learned quite the opposite. Injecting from 330 to 60 deg BTDCf was absolutely killing my top end power. I’m talking 100++ hp loss at 7500 rpm. I got all the power back by clipping EOI at 110 deg BTDCf, which is still much later than the GM crowd believes is good. They say keep all the injection in the intake stroke (>180 EOI).

Anyway, pushing EOI that much earlier really shrunk my injection window and killed the blend, so I had to increase gdi pressure to 3250 psi to get the blend back up. As it stands, I’m bottoming out at 70-75% blend which is still lower than I want. I’ve been kicking around upgrading gdi injectors so I can maintain my 110 deg EOI but get more flow in a shorter period of time, which would allow higher blend at high rpm (when time is limited) and also more pump capacity by lowering its pressure.

So installing both would keep the GDI blend @90% in most situations involving E85?
Situations. It's hard to get answers about how things are done even though DI is very commonplace now.

 

[TX-Ripper]

If E85 is in the mix, Di blend Is irrelevant.

The Coyote engine platform performers phenomenal 100% port injected.

That is the reason you’re not getting answers, no one does it.

 

[illtal]

If E85 is in the mix, Di blend Is irrelevant.

The Coyote engine platform performers phenomenal 100% port injected.

That is the reason you’re not getting answers, no one does it.

cool

 

[engineermike]

While it is true that ethanol direct-injected actually benefits more than gasoline, if you don’t need it you don’t need it. Tx-ripper is right, that if e85 gets you all the knock suppression you need, then why bother doing things to get more?

 

[illtal]

While it is true that ethanol direct-injected actually benefits more than gasoline, if you don’t need it you don’t need it. Tx-ripper is right, that if e85 gets you all the knock suppression you need, then why bother ding things to get more?

Didn't think about it like that. I am a 93 and GT260+ guy but I'll probably need it once i go big boy boost

 

[markmurfie]

Many people know about and discuss DI and its effects on detonation. Consensus is the more fuel injected by the DI injector closer to ignition, the more benefit you get from it. When that problem is solved with the fuels octane this discussion is pointless. You could discuss the negative consequences of too much DI injection when it is not needed.

What not many know is:
DI, when its pulse width is controlled by PI supplementation and timed correctly, does increase an engines VE. The effect is just a few percent, there's different hypothesis about the exact timing, but it is there. It is similar to conversation about benefits from water/methanol injection, when not octane or fuel limited. What size injector and water/meth mix is best for cooling air charge making it denser with out impeding its flow. When you are octane limited, you can't have these discussions about the details. You are discussing how much is it going to take to become not octane limited and you need more VE or boost.

Literally from page 1, what I was trying to discuss, and ProCharger showed their ignorance to the topic.

Fords OEM strategy goes after VE first, then detonation. They could have done the opposite using a split direct injection strategy like GMs. Its a limitation to the pump gas die hards and unfortunate that only have the one option.

 

[engineermike]

Correct on injection timing intake vs compression stroke, Mark. Since I am knock-limited and the air flow rate is governed by the supercharger (rendering VE advantages almost completely moot), I manipulated pressure and SOI to achieve a ~40% of total fuel flow to be injected during the compression stroke. As good as this [maybe] was for knock suppression, it killed the power. I don’t think small VE changes (accompanied by zero airflow difference) was responsible for 100-150 hp loss. More likely there was either a) 40 deg is not enough time for evaporation between EOI and spark, b) poor mixture distribution in the chamber during firing. As I moved the EOI clip earlier in the cycle, I watched power increase drastically. I’m currently running a minimum EOI of 110 btdc and I think it will take a day on a dyno to fine tune it further.

So did I lose any knock suppression by ending injection 50 deg sooner than before? I can’t say for sure yet because the dew point has been 70-75 deg for the last several months, which suppresses knock as well. I had the borderline timing established using a late EOI by finding knock limits at a ~30 deg dew point so I won’t be able to compare until our air dries out some in the fall.

I still think I could benefit from larger di injectors for a few reasons. I could get more di injection in the existing window, lower gdi pressure and get more pump flow while staying in the window limits, and potentially extend the rev range (since the time window shrinks as rpm increases). Initially pressure increases can help this but I’m already at 3250. I just need to figure out how I want to approach the injector situation.

As far as GM stuff goes, I don’t have direct experience but I’ve been told twice by different GM tuners to avoid injection during the compression stroke at all.

 

[markmurfie]

Consider this, Methanol injection on diesel engines, can easily see over 100HP increases with no tune or boost setting changes. This is numbers like 380 to 490, at peak diesel RPMS usually under 5252. Meaning it is a quite significant change in VE. While moving where you inject fuel in the air charge, and the timing of the DI during intake shows small VE changes. The difference between a complete dry air charge and a wet one will be significant.

You don't want to pull too much from the PI shot into a compression biased DI shot to suppress detonation. You will lose significant VE to lower air charge densities. The GM guys don't have any PI shot so when they bias toward compression, they really see the negative affects, and think its the worst strategy anyone can do.

 

[engineermike]

@markmurfie , to clarify, are you saying that 100-150 hp changes resulting from a 50 deg EOI difference are due to changes in VE, even though the airflow was the same?

 

[Andy13186]

I wonder what the max hp on e85 would be with DW400 in tank pump, id1050x and this xtreme DI pump with stock di injectors. I am very interested in this, I am whippled on e85 without a return fuel system currently and I dont want a return style fuel system because of a lot of reasons, fuel heating, or relying on hobbs switches, multiple pumps that could fail, or both being some of the main ones. Just more stuff that could go wrong it seems.

Are new DI injectors required at a certain known hp with this setup?

 

[Angrey]

I wonder what the max hp on e85 would be with DW400 in tank pump, id1050x and this xtreme DI pump with stock di injectors. I am very interested in this, I am whippled on e85 without a return fuel system currently and I dont want a return style fuel system because of a lot of reasons, fuel heating, or relying on hobbs switches, multiple pumps that could fail, or both being some of the main ones. Just more stuff that could go wrong it seems.

Are new DI injectors required at a certain known hp with this setup?

I saw your post in the other thread. While I agree with your philosophy, I think you're trading one evil for another. The better question is to first find a tuner who's done this type of setup and is confident they'll be able to tune it properly.

Relying on a mechanical pump to provide all or most of the required fuel has it's advantages (although it'll be expensive), but it has serious challenges if you read through this entire thread. Not only are you on the hook for the pump (which is pretty spendy comparatively) but depending on when you inject, you may end up having to go bigger injectors (than normal return style port injection) so that's additional cost too. It seems like it has huge potential but a lot of bugs to work through until the industry catches up. Ironically, the big HP competitive 1/2 mile cars all eventually run out of spark and go with a mechanical pump system (driven by a cable). Not DI, but mechanical and directly indexed to RPM.

If you don't want a return style (and you find a tuner that's willing to tune big numbers on returnless) you might look into twin DW440's. Those pumps are brushless, fit inside any hat (42 mm cases) and are true variable speed, can be boosted to 22v (almost doubling their output). With no boost, at 75 psi (net 55 with boost) those will spit out 750 liter/hour. Boosted they'll probably spit out well over 1000 liter/hour which is more than most tuners are ever going to do on a returnless. It's able to be controlled PWM. Again, that particular arrangement is probably novel and a little unique, but it's similar to just using upgraded OEM pumps managed by the computer.

Sponsored