T-leg Piggyback ECU R&D

[jake15]

i got my 440cc injectors today from a V6 STi, and dropped them into some 92-94 N/A rails i had at the shop, and they fit pics of the injectors tomorrow and a write-up when i install them on thursday hopefully

[legacy92ej22t]

Sweet! It looks like we're finally getting some solid injector upgrade options! That's killer.

[azn2nr]

thats the best news ive heard all year

[Nomake Wan]

Hell. Yes.

[simonasaur]

The JDM 2liters came w/ 450is injectors and those drop right in.

[BAC5.2]

What are you using to meter fuel to them? And how are you overcoming the 5-volt Maf reading?

[jake15]

What are you using to meter fuel to them? And how are you overcoming the 5-volt Maf reading?

right now they're just going to be put in. and in a month or two, i'll get a S-AFC hopefully. and eventually after some more susupension upgrades i'll get some type of aftermarket engine management, but thats a LOOONG way off. what do you mean "overcome the 5-volt maf reading?" i did a search, and i kind of understand what your asking, but do elaborate.

[THAWA]

are these any different from the usdm sti injectors? They looked the same as the EJ255's injectors also.

[BAC5.2]

Well, as it would appear, our MAF maxes out it's reading at 5 volts. Roughly 10 to 12 psi on a 16G. I've verified the maxing on the scantool.

This has a problem, because when maxed out, the injectors go 100% idc.

So big injectors + 100% IDC = trouble.

Not to mention that the idle fuel trims for stock injectors are going to make you have a MAD rich idle, as well as injectors that are going to throw more fuel than stock at all levels.

You need to be able to scale the injectors down, without mussing with the MAF signal. But you still run into the problem of the 5 Volt limit.

An SAFC modifies the MAF signal to control fuel, which works (but throws off the load computations, which effect timing). But the injectors still aren't properly scaled, so you just trick the ECU into seeing less air than it's actually getting.

Ideally, you would be able to scale down the injectors, so that the signal going to them is adjusted and the pulsewidths lengthen enough to bring the amount of fuel that the injector is squirting near stock.

With a bigass injector, sometimes they aren't able to meter that little fuel, and the engine will idle like shit. 440 isn't big enough to do that, but anything above 600 or so, and you'll run into a lopey idle. Not a lot of tuners can tune a really stable idle from varying throttle positions (quickly depressing the clutch from high RPM and letting them fall), and not a lot of tuners can tune the roll-off-throttle where the Throttle position sensor snaps quickly to closed (the jerkiness you get when quickly letting off the throttle in gear, is the ECU not being able to properly compensate for that, and cutting fuel quickly).

I know Phil from Element Tuning can tune an idle on 800cc injectors, and can completely tune out the lift-throttle-fuel-cut with his Hydra ECU. It can probably be done with any other user-definable ECU.

[legacy92ej22t]

So big injectors + 100% IDC = trouble.

This is true but with the 440cc I don't think it will be too bad. I was going to run 440cc injectors if I got the EJ20G. If you're running higher boost, like some of us are, then I think it would actually be benificial at high boost. You will run rich which can be bad for performance but not really bad for the engine. It's not like running lean which is *flash*"danger to engine"*flash*. hehe.


It will probably make your idle a little weird and partial throttle performance will probably suffer a little too. If you're running at WOT on high boost it will be sweet though.

[BAC5.2]

But think, that your entire fuel curve will jump up, as the engine will be getting more fuel at ALL RPM's and loads. The ECU will think it's metering X cc's of fuel, but it's really getting Y cc's.

I bet the engine would try to stall every time you quickly let off.

[legacy92ej22t]

Hmm, I'm not sure how it would react when the throttle snaps shut but I bet it'd pull like a bastard at 18 psi on my car!

That's also why I was saying that idle and partial throttle applications will suffer. The only time it will be benificial at all is at WOT with high boost.

If you're like me then there is only two speeds anyways. Too fast or too slow.

[vrg3]

440cc/min happens to be just within the 20% range that the ECU's fuel trims are allowed to max out at. So at light load you'd expect the ECU to be able to figure out how to get somewhat reasonable performance out of these injectors, assuming the oxygen sensor's in good condition. I certainly don't think you'll have problems with stall-outs and stuff.

The weirdest situation will be when you're just above the MAF sensor's range. Almost all sensors max out at 5 volts -- the issue is that 5 volts on our sensor only represents about 200 grams per second, which isn't very much flow. So these injectors would make you run ridiculously rich when on mild boost. But at high boost, like Matt says, it should start to get good.

You'll still be driving the injectors at 100% duty cycle, though, and so shortening their lifespan. But side-feed injectors are inherently more resistant to this kind of damage than top-feed injectors.

[douglas vincent]

Going up to 440 from 370 isnt as bad as I did from the 270 to the 370. But I did/am suffering some herky-jerky problems now, even with my homemade AFC.

[boostjunkie]

So big injectors + 100% IDC = trouble.

It's not like running lean which is *flash*"danger to engine"*flash*. hehe.

It will probably make your idle a little weird and partial throttle performance will probably suffer a little too. If you're running at WOT on high boost it will be sweet though.

Better seafoam that b!tch frequently, along with change the plugs! LOL!

*imagines Matt going WOT whereever he goes*

In all seriousness, no one will really know how the car will drive with the larger injectors. They are "fairly" close to the stock size that the idle/low boost characteristics might be "liveable."

Your MPG will probably suffer tho, and if I remember my old car, it was already hard on the gas.

[jake15]

Ok, so i can put these in and worry about slightly less gas milage? if i keep the boost around 12 psi the gas milage will increase because the injectors will not be running at 100% right? i'm at just below fuel cut right now, and i really dont have any plans to turn up the boost anymore in the near future. so would i get reasonable gas milage at 11-12psi with the 440's? i understand the 5 volt thing now though thanks for your explanations guys

[vrg3]

The injectors will be running at 100% duty cycle as soon as you exceed the MAF sensor's signal range. With a TD05 that seems to happen somewhere between 10 psi and 12 psi, depending on the rest of the system.

The stock EJ22T injectors should flow enough fuel for 11-12 psi with your setup, so you might want to reconsider your injector swap. Get a wideband oxygen sensor setup instead so you can actually find out what's going on.

In almost all cases, you will not get better mileage by putting the wrong injectors on the car. You'll almost certainly get worse mileage. In this case, though, if you're lucky, it won't be much worse.

[georryan]

After reading all this, I really am starting to think that is it worth the money to buy Kelley's ecu and change our whole setup to a MAP(?) setup rather than MAF.

[jake15]

how much do wideband 02 sensors cost? i think i'll put the injectors in on saturday and see how it goes for a couple weeks. if i dont like them, i can always switch them back

[BAC5.2]

A wideband O2 sensor is right around $300USD.

[BAC5.2]

After reading all this, I really am starting to think that is it worth the money to buy Kelley's ecu and change our whole setup to a MAP(?) setup rather than MAF.

Yea, MAP can certaintly be beneficial, but a MAF setup is fine, you just need to redefine how much air each incriment of the MAF sensor means.

On the stock ECU, it maxes at 5 volts, which represents 200grams per second to the ECU. So when the ECU reads 5 volts, you are pulling in 200 grams per second (that the ECU can see).

So I would imagine, that if you modified that, and told the ECU that at 2.5 volts, 200 grams per second was flowing, you could double the range of the MAF sensor without really adversely effecting load calculations or anything.

So really, ANY engine management that can independently control fuel and timing, will be able to fix this problem.

Vikash would probably be able to confirm or deny this.

[vrg3]

Right.

But you'd need to actually have a MAF sensor that had a transfer function that made 200 g/s come out as 2.5v.

As a bit of trivia, if you actually had a MAF sensor that read 200 g/s as 2.5 v, it'd probably top out around 800 g/s, since most MAF sensors have quadratic transfer functions.

[-K-]

Well I think the best way to fix this 440cc injector problem is find the right MAF to use with it.
Find a MAF that hits 5v at 235-240 grams per second. Then you dont have to scale anything.

[BAC5.2]

Vikash - So how would I be able to adapt to a MAF that can read levels like that?

When I go to aftermarket engine management, what will I need to do to get it to read more than 200 grams per second from the MAF?

[vrg3]

If you're using aftermarket engine management then you're almost certainly using something based on a MAP sensor, so this isn't really an issue.

But with the stock ECU, there are two approaches you could take. You could try -K-'s suggestion, using a higher-range MAF and larger injectors and trying to balance the two.

The "right" way to do it though would be to reprogram the ECU with the transfer function for the new MAF sensor.