High Revv NA Setup?

[Coheed]

Originally Posted by serialk11r

The TGVs are operated by electronic servos. The plastic thing splits the intake port into 2 halves, one of which is blocked off by the TGV. Can't throttle an engine with that.

Why not? Don't split the port and run them like ITBs. you can modify the servo opening rates in the ECM to generate your own tumble at low rpms. Problem solved. Or run dual servos on each side, one as the main "throttle" plate and the other as a TGV.

Or don't use the TGVs as ITBs, but rather use ITBs to generate tumble like the TGVs do in the first place. That's all I'm saying. There are a lot of different ways to tumble the air, or change intake velocity.

[serialk11r]

Originally Posted by Coheed

Why not? Don't split the port and run them like ITBs. you can modify the servo opening rates in the ECM to generate your own tumble at low rpms. Problem solved. Or run dual servos on each side, one as the main "throttle" plate and the other as a TGV.

Or don't use the TGVs as ITBs, but rather use ITBs to generate tumble like the TGVs do in the first place. That's all I'm saying. There are a lot of different ways to tumble the air, or change intake velocity.

ITBs don't generate tumble. You block off half the intake port so that all the air is going through the valve on one side, generating tumble flow in the cylinder. You need the wall thingy splitting the port because you can't place a valve right over the port. A throttle body upstream doesn't do the same thing. If you try to get rid of the TGV low end combustion quality will be complete crap. If you put a second valve over the TGV assembly to block the whole intake port, sure you can throttle it like that, but adding that would be even more expensive than a separate ITB setup.

[Dimman]

TGVs are irrelevant flow restrictions in a D4-S system. Turbo conventional DI (Subaru's version for the Legacy and ?) needs tumble. D4-S spray patterns vastly reduce the need for low rpm charge motion to produce good mixture/combustion.

So a high flow/rpm head on this should still have acceptable lower rpm response, appropriately tuned.

[Coheed]

Originally Posted by serialk11r

ITBs don't generate tumble. You block off half the intake port so that all the air is going through the valve on one side, generating tumble flow in the cylinder. You need the wall thingy splitting the port because you can't place a valve right over the port. A throttle body upstream doesn't do the same thing. If you try to get rid of the TGV low end combustion quality will be complete crap. If you put a second valve over the TGV assembly to block the whole intake port, sure you can throttle it like that, but adding that would be even more expensive than a separate ITB setup.

True, a conventionally designed ITB is designed NOT to tumble the air. But that doesn't mean I couldn't make it do so.

I get what you're saying, but the FA20 has an inefficiency region that may take unconventional methods to solve. It would be nice if we could keep it simple. Perhaps a dual runner intake would be a better option. Run one set of runners with the wall built into it, with the shorter runners without wall or restriction for the top end.

Low end combustion on the FA20 is good. Mid range is where there is an issue.

You do bring up a good point about cost. The engineers for this engine are going to have to make the accountants happy too. Which begs the question, why add anything at all? You don't NEED the TGVs to solve the torque dip. After all, there's lots of cars out there with no torque dip and no TGVs. Right?

ITBs would be cool, but they aren't likely to happen on production engines in this price range. I'll guarantee that.

Perhaps all this engine needs is a better intake manifold design.

[serialk11r]

Originally Posted by Coheed

Low end combustion on the FA20 is good. Mid range is where there is an issue.

The torque dip is not because of combustion efficiency. There is no combustion efficiency issue in the midrange. Combustion efficiency is basically only a problem at low speed.

Currently, low speed combustion is okay because there is the D4-S system, and because the ports are designed to make it work. Low load the direct injectors do their stratified charge thing, high load port injectors supply some fuel because the DI can't mix the fuel well alone. Subaru wants to ditch this setup but keep direct injection, which is why they are bringing TGVs in. Direct injection requires high tumble flow to work, but a high tumble port restricts airflow. The TGV allows good top end power when open, good low end charge motion when closed.

@Dimman, arghx7 said the ports on this are not what you'd call high flow ports. He said D4-S just allows you to strike a better compromise than the usual high tumble port + DI combo. Remember that the direct injectors are providing all the fuel below some load level.

[Coheed]

Oh ok, got what you're saying. I didn't know they were deleting the port injectors next year. I figured they were adding the TGV to alleviate the torque dip issue.

[serialk11r]

Originally Posted by Coheed

Oh ok, got what you're saying. I didn't know they were deleting the port injectors next year. I figured they were adding the TGV to alleviate the torque dip issue.

I don't remember what those pictures and stuff actually were, and I don't know if that's actually the plan for next year, but it seems like a plausible route for them to take. I think the original thread it appeared on said something along the lines of "these are what the plan is for the next FA20" but I don't think it was totally certain. If they do pull this off that would be pretty neat, since it's been a while since anyone's done a "dual intake runner" (not exactly, but it's got a valve and 2 air passageways lol).

[Dimman]

Originally Posted by serialk11r

I don't remember what those pictures and stuff actually were, and I don't know if that's actually the plan for next year, but it seems like a plausible route for them to take. I think the original thread it appeared on said something along the lines of "these are what the plan is for the next FA20" but I don't think it was totally certain. If they do pull this off that would be pretty neat, since it's been a while since anyone's done a "dual intake runner" (not exactly, but it's got a valve and 2 air passageways lol).

They are from the new turbo Legacy.

The high flow I mention is for a modded head. The D4-S reduces the need for turbulence/velocity, so the modded head may get a couple thousand rpm more on the lower end of its powerband than what could typically be expected.

[brillo]

I thought the dual injectors were also to help keep the valves clean as well as for power and efficiency. Subi wants to dump the port injectors completely? Do the direct injectors have the volume capacity to feed the engine across the rev range?

[Nafe]

Originally Posted by noAE86

HAHAHA so we are going back to old school toyota TVIS systems, but instead of vaccume operated its now electro servo,
(granted one is for tumble and one is for verlocity but still..)
it makes the car feel THAT much closer to the AE86, i love it,

(i have about 6 sets of 20valve blacktop quads and 4 sets of silvertop quads and 3 sets of SR quads in my shop if anyone is wanting quads, they sell for about $100 N.Z$ so about $60 U.S,

Hey mate, I don't suppose you have a set of honda beat triple throttles lying about?

[arghx7]

Originally Posted by Turdinator

ITBs are always expensive. In fact, as has been mentioned several times before, NA mods are always expensive for the HP gains. I don't see why they'd be hard to control as long as you are using a suitable ECU.

I can't think of an aftermarket racing-type ECU that can drive two electronic throttle valves on their own, without some kind of piggyback setup. If you find one under $5000, let us all know.

I was thinking last night that perhaps a twin throttle setup would be easier than ITBs.

They're actually very similar to ITB's in terms of the way they are physically driven. On ITB's for electronic throttle, you have all the throttle valves for a bank driven on a common shaft. There has to be software learning to calculate and compensate for differences in airflow between cylinders. So if you have two banks with ITB's, you have two electronic throttles essentially. In one sense that's the same as two upstream throttle valves.

Some engines have an upstream throttle before the plenum and then ITB's right at the intake port, like the Lexus LFA engine.

We could use two stock throttle bodies and just splice the signal from the ecu to each in parallel. Although i don't know if that electronically feasible.

Never say never I guess, but you do realize that each electronic throttle module has two throttle position sensors on it used for feedback? So how is the computer going to compensate for variations in throttle valve deposits if you are feeding it a dummy signal to drive the DC motor? The electronic throttle learning won't be accurate for the other throttle. I suppose that matters a lot less in a racing environment with very little part-throttle driving, but I just don't see it being feasible on a street car.


Nobody has addressed my point about the lack of variable valve lift. On Celica 2ZZ or basically any Honda engine, you have more than one lift profile. That allows you to have some semblence of low speed torque. If you can succeed in going with a very aggressive camshaft design you will lose a lot of your low-end torque on this engine.

[arghx7]

Originally Posted by brillo

I thought the dual injectors were also to help keep the valves clean as well as for power and efficiency. Subi wants to dump the port injectors completely? Do the direct injectors have the volume capacity to feed the engine across the rev range?

The valve deposit thing is simply a welcome side effect of the dual injection. The main value in dual injection is to help with some of the compromises you have to make when designing a direct injection combustion system. The port injectors are also much better for particular emissions, and particulate emissions standards are about to start tightening... you may be seeing particulate filters on gasoline engines in a few years.

[Turdinator]

Originally Posted by arghx7

I can't think of an aftermarket racing-type ECU that can drive two electronic throttle valves on their own, without some kind of piggyback setup. If you find one under $5000, let us all know.

As my car is still 3 months away from delivery i haven't really researched this, so i bow to your greater knowledge. I was under the impression Motec and Autronic had electronic throttle capabilities. Although they may not fall under your $5k threshold.

Originally Posted by arghx7

They're actually very similar to ITB's in terms of the way they are physically driven. On ITB's for electronic throttle, you have all the throttle valves for a bank driven on a common shaft. There has to be software learning to calculate and compensate for differences in airflow between cylinders. So if you have two banks with ITB's, you have two electronic throttles essentially. In one sense that's the same as two upstream throttle valves.

Never say never I guess, but you do realize that each electronic throttle module has two throttle position sensors on it used for feedback? So how is the computer going to compensate for variations in throttle valve deposits if you are feeding it a dummy signal to drive the DC motor? The electronic throttle learning won't be accurate for the other throttle. I suppose that matters a lot less in a racing environment with very little part-throttle driving, but I just don't see it being feasible on a street car.

With a twin throttle body setup you could us a balance tube between the manifolds to over come any part throttle imbalance. Or just position the throttles in such a way that you could run them off the one electronic motor.

I hadn't thought about the feedback loop that ETB must run. It certainly kills my idea of splicing the signal.

Originally Posted by arghx7

Nobody has addressed my point about the lack of variable valve lift. On Celica 2ZZ or basically any Honda engine, you have more than one lift profile. That allows you to have some semblence of low speed torque. If you can succeed in going with a very aggressive camshaft design you will lose a lot of your low-end torque on this engine.

The simple answer is that changing the cams is a compromise of greater performance at the expense of easy drivability. However a properly set up engine won't lose too much drivability as long as you don't go too insane with your cam choice. 304* will be much harder to drive in the low range than 272* but the variable timing will make both cams far more livable then and engine without variable timing.

If you are worried about low end torque then keeping the FA20 NA probably isn't for you.

[vignesh]

Our cars has some potential

[ame="http://www.youtube.com/watch?v=OutWSkfaNEQ"]http://www.youtube.com/watch?v=OutWSkfaNEQ[/ame]