Which injectors, fuel pump, injector control?

[free5ty1e]

Hey...

Do we know of any larger direct replacement injectors (such as from a WRX or an STi) that will flow enough in our cars to support upwards of say, 20psi of boost? Which fuel pump would support this -- i'm thinking something from Walbro but dont know which model would be the best to select based on ease of replacement or installation and flow rate.

Also, do EJ22t's run stock on a fixed-pressure fuel system? 3 bar? 4 bar? Or do we have rising pressure as the boost increases? Would any of this have to change to support those high boost levels?

One more thing... what are the tested options for standalone injector control/engine management? I know the stock ECU will probably not like the larger injectors and more flow from the pump, unless I can figure out a way to reprogram it - which doesnt seem like it will happen anytime soon...

Any input would be appreciated.

[vrg3]

If you do a search you can read about the injectors question and the engine management question, which have been discussed in several other threads. Try including "nissan" in your search query about injectors.

Fuel pumps interchange fairly easily, but I believe the Walbro GSS341 is the one that's meant to fit our cars best.

The fuel system maintains a constant pressure differential of 36.3 psi between the fuel rail and the intake manifold. That means that at 0 psig manifold pressure, fuel pressure is 36.3 psig, and at 10 psig manifold pressure, fuel pressure is 46.3 psig, and at -10 psig manifold pressure, fuel pressure is 26.3 psig.

You would certainly have to use a higher-flow pump for high boost levels, but you may well do okay with the stock fuel pressure regulator and rails. The boostjunkie Legacy was running around 17 psi of boost with the stock rails and regulator.

[free5ty1e]

Thanks again, V... that makes sense. Didn't even know differential fuel pressure regulators existed. Keep in mind here my previous car experience was pretty much limited to Saturns

Just out of pure curiosity, how far does that constant pressure differential go? I mean... at 20psi would it have any problems maintaining the 56.3psi fuel pressure, assuming the higher flow pump was in place?

[vrg3]

Are you saying Saturns don't use this system!?

I thought almost all modern port fuel injection systems did. The idea is to maintain a constant pressure differential so that the quantity of fuel injected is purely a function of how long and how fast the injector is opened.

The way the FPR works is by having fuel press against a springloaded diaphragm. On the other side of the diaphragm is an air chamber at manifold pressure. When fuel pressure pushes against the diaphragm, it opens a valve, allowing it to flow back to the tank. The preload on the spring is what determines the static fuel pressure.

Since neither side of the FPR is actually exposed to atmosphere, I don't think the FPR "cares" about how high manifold pressure actually is; in fact, it doesn't "know" how high manifold pressure is. So, it seems to me that as long as the pump is up to the task of creating the fuel pressure, the FPR will regulate it.

Note that there could be a potential problem at low manifold pressures if you used a really really powerful fuel pump -- in principle the valve in the FPR could be unable to open far enough to bleed off all the pressure being built up. This is a problem some RS-T people run into: the FPR installed on newer non-turbo Subaru engines doesn't flow very well, so a Walbro pump can cause overly high fuel pressure at idle on new RSes. I haven't ever heard of this being an issue with Legacy Turbo FPRs though.

[free5ty1e]

Heh... nope, my Saturn didnt even have a return line. Just a constant ~45psi of pressure, on 190cc injectors. The earlier ones had a return line and all but I think it was still fixed. I could be wrong... but none of them came with turbos

Anyway, once again my understanding has been increased by your informative posts... heh... thanks!

[vrg3]

Heh, no problem...

So the Saturn's ECU has to take into account the fact that the injectors will flow better at idle than at WOT... interesting.

Hmm, now that I think about it, I guess it makes sense that a returnless fuel system wouldn't have manifold-referenced fuel pressure, since pressure is regulated far away from the manifold. Anyone know if other returnless systems are like the Saturn's?

[free5ty1e]

heh... Saturns have some fussy PCMs. I can attest to that. The 190cc/min injectors they have can react to some small pulse widths for idling. And yes, you're correct - the FPR on my later model was actually in the same unit as the 100,000 mile fuel filter, both located directly underneath the gas tank. The older models had FPRs on the rail, with a 24,000 mile filter right on the firewall. Ah, more useless knowledge at its best

Anywho, a thought:

What are the realistic pressure/flow limits of our stock injectors? Theoretically, since they just open and close with the pulse width put out by the ECU, couldn't they just flow more and more with the higher fuel pressure levels as manifold pressure increases? Say... at 20psi (man I just love that number), when the fuel pressure regulator would be allowing 56.3psi of fuel pressure to the rail from the fuel pump (upgraded to the Walbro of course), wouldn't the injectors be flowing the proper amount to compensate for the increased boost already? Couldn't the ECU read the O2 and adjust the PW to provide the proper a/f ratio for 20 psi, assuming the fuel pump could support it? ....hmm, assuming as well the fuel lines could also hold that pressure without bursting.

Or do our injectors sort of crap out at a known pressure threshold or something? If not, I can't see a reason to upgrade anything but the fuel pump and then just crank the boost, as dangerous as that sounds

[vrg3]

In the stock configuration, the injectors don't flow more and more as manifold pressure increases. They flow the same amount, regardless of manifold pressure.

The reason the FPR needs to increase fuel pressure on boost is just to keep up with the boost. If it didn't, the boost would push against the fuel that the injector's trying to inject!

You could make them flow more and more as boost rises if you increased fuel pressure by more than a 1:1 ratio, which is what a rising rate fuel pressure regulator does.

[evolutionmovement]

The excess pressur will only put out so much more fuel - you still have the mechanical restrictions of the stock system. Also to use the reactive compensation of the o2 sensor would not be an optimal way of running. Someone with more tuning experience would be better qualified to answer, though.

Steve

[mTk]

Any idea on what the limit would be using a rrfpr?

MK

[vrg3]

Relying on the oxygen sensor stuff isn't really an issue... if the ECU could provide enough fuel to run rich at 20 psi it would. But with any decent turbo, you'll hit 100% injector duty cycle before you hit 20 psi.

I don't know what the feasible limits are with an RRFPR... But fuel flow only goes up with the square root of fuel pressure, so you kind of get diminishing returns.

In any case, though, if you're going to try to make the kind of power it sounds like you're shooting for, you really should use programmable engine management that allows you to tune at those levels. Control is key, and you can't have it with either an ECU running far outside its maps and/or a rising rate fuel pressure regulator.

[rsstiboy]

i go no where near 100%duty cycle on mine, I have a walbro 500HP fuel pump, 500cc injectors, and running 19psi on my TD05H, my fuel pressure is 45psi rising to 65psi.

[free5ty1e]

hmm... I see. That makes sense. Didn't consider that the manifold pressure would be fighting the incoming fuel spray if the pressure didn't increase. Can our fuel lines can hold the high pressures an RRFPR would be putting out at those boost levels, or are we looking at running new lines and everything?

RSSTIboy - you're not actually driving a Legacy though are you? Might I inquire as to which injectors you've got, and which fuel rail? And are you using the stock ECU to control the injectors, or what?

[vrg3]

Yes. You have done the right thing and upgraded your engine management systems.

But you use an RRFPR, right? I don't understand that decision.

[free5ty1e]

Yeah... when I was planning the turbo Saturn's higher boost levels, especially since it didnt have a return line (so RRFPR was not an option and not wanted anyway) was to get much higher flow injectors (and fuel pump of course) and just control the pulse width finely enough to maintain idle and drivability at the stock 45psi fuel pressure, while providing all the fuel I'd need at full boost.

Thats still the best idea here right? RRFPRs were nothing but trouble as far as I could tell from the earlier model turbo saturn owners I spoke with that experimented with them.

[rsstiboy]

RSSTIboy - you're not actually driving a Legacy though are you? Might I inquire as to which injectors you've got, and which fuel rail? And are you using the stock ECU to control the injectors, or what?

I am in australia, I have a EJ20t motor in my RS liberty, I am using possumlink engine management, and I am not sure what injectors they are, all I know is that they are bigger than 500cc, they came on a JDM wrx halfcut I bought. I am using standard fuel rails, but they are both right hand side rails(the ones without the fpr) thats why i have a rrfpr on the car.

[vrg3]

I don't understand how having to use an aftermarket FPR explains why you use an RRFPR... Wouldn't it make more sense just to use a standard 1:1 fuel pressure regulator so that tuning the ECU would be easier?

[rsstiboy]

tuning is no different than a normal fpr, think about it, when boost comes on so does fuel. its only when you start playing around with boost after your tune that it can run differently.

[vrg3]

But if the rise ratio isn't 1:1, then a 5ms injector pulse at 10 psi would inject less fuel than a 5ms injector pulse at 15 psi...

[rsstiboy]

but you tune your load points based on airflow and load whilst watching your air/fuel guage.

[vrg3]

Okay... That makes sense, and if you're tuning based on the instruments' reading I guess it shouldn't really matter.

I guess I was thinking in terms of the only car that I've ever tuned engine management for, and it was a speed-density system so the fuel map was actually a volumetric efficiency map. A rising rate regulator in that case would have thrown off the perceived volumetric efficiencies. But in your case you have an airflow meter.

[rsstiboy]

I got rid of the airflow meter, I only have a map sensor and oxygen sensor, then I have another oxygen sensor wired up to my air/fuel gauge ( you know the ones with the dancing lights that aren't accurate) I used a wideband O2 sensor for tuning.

[vrg3]

Oh... I thought you were running a Z32 MAF. That was before, I guess?

Yeah, now that I think about it, the only "difficulty" with tuning an RRFPR-equipped speed-density car is that you can't just use volumetric efficiency to determine the fuel maps. But if you're tuning entirely empirically anyway, that makes no difference.

When we tuned my brother's car we started by looking at torque curves for similar engines and came up with our initial VE table based on that.

Out of curiosity, how come you don't just run with a wideband all the time? The sensors aren't that expensive, and it seems reasonable considering the amount you've invested in the car.