2014 Subaru BRZ - Minor and Major Mechanical Changes

[Snoopyalien24]

So, this is our "mule in camo" till next year?

I love it. Thanks WoW. Deff will be waiting this out to see what happens

[serialk11r]

Originally Posted by Dimman

For reference on the 'old' Toyota motors the IS250 with direct only had TGVs and ACIS (variable length intake tube), but the IS350 with the D4-S has neither.

Did they think the smaller 2.5l V6 needed more help than the 1000cc larger motor?

Think about it. That's exactly what's going on, the 2.5L has direct injectors only, so it needs TGVs if it's going to both make power up high and have reasonable low end combustion stability.

On the 3.5L motor, they have port injectors to "limp" around at low rpm, that's how I think of it. Except the only thing is the port + direct happens to give a slight improvement over either alone, so you can't really call it limping.

[Dimman]

Originally Posted by serialk11r

Think about it. That's exactly what's going on, the 2.5L has direct injectors only, so it needs TGVs if it's going to both make power up high and have reasonable low end combustion stability.

On the 3.5L motor, they have port injectors to "limp" around at low rpm, that's how I think of it. Except the only thing is the port + direct happens to give a slight improvement over either alone, so you can't really call it limping.

I've been going over one of the 2GRFSE pdfs posted (probably by arghx7). In the SAE paper they tested with and without the swirl control valves and found they cost 20% of volumetric efficiency at 6400 rpm, but gain 4% more torque at 2800 rpm from better combustion even though the 'open' motor had slightly higher VE at the low rpm.

20% VE penalty at 6400 rpm (probably even worse as rpm increases) for 4% more torque at 2800 rpm?

Fuck. That. Shit.

No way SCV/TGVs are going on this motor.

(If the .pdf isn't in arghx7's D4-S stickied thread, let me know and I'll attach it. Title is 'Toyota 2GR-FSE GDI PFI engine paper.pdf'

Edit: it's in his sticky. IS350 section.

[serialk11r]

It's not about the peak torque, it's that combustion stability is affected across the board so you'll probably have rough idle and misfires or significant torque fluctuations, all that shit, if I understand correctly.

arghx7 indicated that the ports on D4-S engines are likely a compromise between tumble and flow, because you really do need the tumble at low speed for the direct injection to work. The port injectors give you more room in terms of how much tumble you need.

Basically, you're kinda forced to have a flow compromise of some sort, and I have this gut feeling that TGVs are the better option. Now of course BMW and Porsche and whatever don't seem to have issues with this, but arghx7 makes it sound like their stuff is much fancier and more capable, so I'll believe that for now even though I don't quite understand how it works. For Porsche I think reduced valve lift increases tumble flow or something with the right design, so they're okay. Not sure how BMW deals with high/low rpm compromise on their S-- engines with only double VANOS, but I'd guess their injector placement and firing have something to do with it. EDIT: wait, BMW S-- engines don't have direct injection? Okay that clears things up.

[Dimman]

Originally Posted by serialk11r

It's not about the peak torque, it's that combustion stability is affected across the board so you'll probably have rough idle and misfires or significant torque fluctuations, all that shit, if I understand correctly.

arghx7 indicated that the ports on D4-S engines are likely a compromise between tumble and flow, because you really do need the tumble at low speed for the direct injection to work. The port injectors give you more room in terms of how much tumble you need.

Basically, you're kinda forced to have a flow compromise of some sort, and I have this gut feeling that TGVs are the better option. Now of course BMW and Porsche and whatever don't seem to have issues with this, but arghx7 makes it sound like their stuff is much fancier and more capable, so I'll believe that for now even though I don't quite understand how it works. For Porsche I think reduced valve lift increases tumble flow or something, so they're okay. Not sure how BMW deals with high/low rpm compromise on their S-- engines with only double VANOS, but I'd guess their injector placement and firing have something to do with it.

Read the paper. The whole point of the D4-S was to break the two compromises. 20% VE penalty over 6400 rpm has no business in a performance motor.

[OrbitalEllipses]

TGVs are not OEM ITBs...

[serialk11r]

Originally Posted by Dimman

Read the paper. The whole point of the D4-S was to break the two compromises. 20% VE penalty over 6400 rpm has no business in a performance motor.

I did, and I asked arghx7 about it, and he confirmed my suspicions. Direct injection simply needs some sort of charge motion to get it working well, port injectors are less picky since you can just vaporize the fuel on the valve. The port injectors aren't supposed to run big duty cycles though, so that suggests they still needed to design some degree of tumble flow into the ports, however that works.

The paper didn't say D4-S was to break the compromise, rather it described the flow compromise that a DI engine sees, the efficiency disadvantage that port injection has, and then they went on and talked about their new engine, without actually saying what they did to the ports.

[Dimman]

Originally Posted by serialk11r

I did, and I asked arghx7 about it, and he confirmed my suspicions. Direct injection simply needs some sort of charge motion to get it working well, port injectors are less picky since you can just vaporize the fuel on the valve. The port injectors aren't supposed to run big duty cycles though, so that suggests they still needed to design some degree of tumble flow into the ports, however that works.

The paper didn't say D4-S was to break the compromise, rather it described the flow compromise that a DI engine sees, the efficiency disadvantage that port injection has, and then they went on and talked about their new engine, without actually saying what they did to the ports.

They didn't do anything to them. Basically they were normal without SCVs choking them and the cutouts that increase tumble in the 3GRFSE.

The purpose of the Yamaha developed double fan spray pattern is

to promote a homogeneous mixture at lower engine speeds without any devices to generate intense air motion to improve combustion stability.

With the developed spray, this test engine can improve at all engine speeds even though tumble is weaker without a SCV.

[serialk11r]

Hmmm...
I guess exactly what "weaker" means is still unknown, we'll see. arghx7 seemed to think that there is some intended tumble, and he knows his s*** lol.

[Dimman]

Originally Posted by serialk11r

Hmmm...
I guess exactly what "weaker" means is still unknown, we'll see. arghx7 seemed to think that there is some intended tumble, and he knows his s*** lol.

There has always been intended motion, even back in carb days. Swirl or tumble doesn't really matter so long as it is mixing the fuel and air.

On my quest for missing torque I fixated on 4800ish rpm because based on valve estimates and piston speeds the intake air speed guestimate is the traditional point where combustion efficiency and flow efficiency intersect which makes peak torque. What the TGV/SCVs do is increase that speed at lower rpm to try for better low rpm combustion.

It's not that inertially tuning is bad, it's that the practical implementations (20% VE loss) are horrendous.
The concept itself goes to multi barrel carbs with vacuum activated secondaries.

For the 'right' way, some sport bikes use the same idea (ECU controlled secondary throttles) but in a vastly more efficient (and expensive) way.

[WingsofWar]

Originally Posted by Dimman

I've been going over one of the 2GRFSE pdfs posted (probably by arghx7). In the SAE paper they tested with and without the swirl control valves and found they cost 20% of volumetric efficiency at 6400 rpm, but gain 4% more torque at 2800 rpm from better combustion even though the 'open' motor had slightly higher VE at the low rpm.

20% VE penalty at 6400 rpm (probably even worse as rpm increases) for 4% more torque at 2800 rpm?

Fuck. That. Shit.

No way SCV/TGVs are going on this motor.

(If the .pdf isn't in arghx7's D4-S stickied thread, let me know and I'll attach it. Title is 'Toyota 2GR-FSE GDI PFI engine paper.pdf'

Edit: it's in his sticky. IS350 section.

Yet those photos i posted of TGVs are on a current FA20 variant..its right there with the FA.

But i get what your saying....sacrificing 20% VE at high-speed for only less than 5% better low-speed stability in VE and TQ. Not a good trade off..aka FUCK THAT SHIT!

Assuming those are similar characteristics on the FA20...how would we improve this? how can we get better VE on all engine loads...better TQ...better low-speed tumble...and not prematurely choke the powerband?

WRX owners sure didn't see any benefit to it...there are TGV delete kits everywhere. But then...where is the data sheet explaining why that particular boxer engine needed tumble?

If we did have TGV in the future...should we get better runners? polished and oversizd? thinner and smoother butterfly valves with low profile pivot rods?

I remember seeing a video looking the stock '09 Nissan GTR throttle body. And admiring how big and beautiful it was...then the guy from Mines told us it was inefficient...and then showed their part..which was only stock but ported and polished which they called oversized...and had their own "high-speed" butterfly valves. I thought that shit was snake oil....then they showed them on a flow-bentch with data...15% better flow from stock. Christ...i didn't know small adjustments like that could change the airflow characteristics so much.

If we look at it that way...could this also change in our application in regards to TGV? Improving tumble without sacrificing high-speed VE?

[Dimman]

Originally Posted by WingsofWar

Yet those photos i posted of TGVs are on a current FA20 variant..its right there with the FA.

But i get what your saying....sacrificing 20% VE at high-speed for only less than 5% better low-speed stability in VE and TQ. Not a good trade off..aka FUCK THAT SHIT!

Assuming those are similar characteristics on the FA20...how would we improve this? how can we get better VE on all engine loads...better TQ...better low-speed tumble...and not prematurely choke the powerband?

WRX owners sure didn't see any benefit to it...there are TGV delete kits everywhere. But then...where is the data sheet explaining why that particular boxer engine needed tumble?

If we did have TGV in the future...should we get better runners? polished and oversizd? thinner and smoother butterfly valves with low profile pivot rods?

I remember seeing a video looking the stock '09 Nissan GTR throttle body. And admiring how big and beautiful it was...then the guy from Mines told us it was inefficient...and then showed their part..which was only stock but ported and polished which they called oversized...and had their own "high-speed" butterfly valves. I thought that shit was snake oil....then they showed them on a flow-bentch with data...15% better flow from stock. Christ...i didn't know small adjustments like that could change the airflow characteristics so much.

If we look at it that way...could this also change in our application in regards to TGV? Improving tumble without sacrificing high-speed VE?

Covered in post above. Sportbikes have the solution. But is it cost effective?

Edit: Question is rhetorical. I'm basically describing 8 ECU controlled ITBs!

[Bristecom]

If there's something strange, in the power band. Who you gonna call? Yamaha!

This is an interesting thread. Keep it up.

[serialk11r]

Originally Posted by Dimman

Covered in post above. Sportbikes have the solution. But is it cost effective?

Edit: Question is rhetorical. I'm basically describing 8 ECU controlled ITBs!

In this case it'd only be 5 though right? Besides, 4 of them would be linked together anyways...You might not even need to vary their position against load, just engine speed.

WoW, 15% seems a little sketch but if it's just the throttle, that could be possible. I am not informed on this by any means but it sounds to me like considering the flow of the entire system rather than just one piece would make that difference look much smaller. With TGVs sitting in a narrow part of the intake tract though, yea I imagine smoothing out the valves and stuff would help. Just throwing ideas out there, but maybe sanding the pivot part a little, and then using some epoxy or foam or something to slightly fill in the small gap, maybe rounding off the edges too (although that could mess up very low load behavior).