Originally Posted by KR-S 86 (Post 2491883)

Anyone have that video of the 9000 RPM turbo FA20?

Originally Posted by FRS Justin (Post 2491847)

I believe the 9k motor was a turbo motor. With stock cams 9k isn't going to do much.

Originally Posted by arghx7 (Post 396698)

Stop comparing this engine to a Honda engine. It has a fixed valve lift. That requires big compromises. And ITB's will be very expensive and hard to control with electronic throttle.

Originally Posted by mad_sb (Post 727334)

7800 was a no brainer for 4B11t's.
This motor can rev, but some of the ecu tables will need to be re-scaled and it will need either cams or FI to make it worth reving deep.

Originally Posted by MattR (Post 390665)

I am also wanting to keep the engine N/A!
While definitely not the end-all factor, Rod/Stroke will also play into high-revving plans. Consider what seems to be a short rod length @ 130-131mm*. This will put your R/S ratio at 1.51-1.52. Not exactly the best starting point for a 9k engine.
A higher R/S ratio will lessen the side-load on your cylinder walls which is nice when wanting to rev to the moon
A few mentions of Honda engines. Hondas are already at an advantage for increased revs due to this factor. K20A engines were in the neighborhood of 1.62, iirc. B16A() screamers were at 1.74.
The mentioned S2000 F20C engines were at 1.82, but note that when Honda introduced the F22C for the S2000, they increased the stroke which lessened the R/S ratio to 1.65. No coincidence that they also decreased the redline.
One of many factors to consider when revving high. Looking forward to fellow all-motor builds!
*If someone knows the actual rod length measurement of the FA20 please educate me. I've been looking all over the damn place and can't seem to locate it.
:happy0180:

Originally Posted by serialk11r (Post 391340)

Oh dear, that's going to be hard. Do cheat with E85 to get you that extra 10% :P Moving the power peak from 7k to 8k with longer duration cams and a different intake manifold (full exhaust and the rest of the jazz too) can net perhaps 8/7*165?lb-ft/150lb-ft * 200hp=251.4hp, at the crank, with very little room to play with after 8k, and a rather peaky powerband before that. From the exhaust numbers and stuff we're seeing, I think 170lb-ft shouldn't be too hard to hit. But now you are looking to find that last 10-15% "drivetrain loss", that's going to be tough on pump gas! (let's admit it, 9000rpm with a 1.5 rod stroke ratio isn't so good for the cylinder bore or bearings) E85 should get you there though, if you can direct inject all of it. When done right the charge cooling effect on pre-combustion pressures alone should be worth 3% more power, or something like that (I did some calculation a long time ago).
@MattR, I think Crower or someone did post a picture of the EJ rods next to the FA rods, and after I asked what the rod lengths were they did post it. I just remember it was 130+/-1 mm.
Actually, this should be much easier than you think, given that we're willing to accept slight compromises in cam design. I believe one of the biggest reasons for the poor low end performance of big cams is that the big amount of overlap introduces large amounts of exhaust gas into the charge, completely ruining combustion efficiency. They say direct injection helps to increase the tolerance for internal EGR, but when you're idling and the vacuum is high, any overlap is going to mean a lot of exhaust sucked back in.
Larger cams however will lose volumetric efficiency at low rpm cruising conditions, which will decrease pumping losses and allow a cooler charge which further increases efficiency.
If you see some Toyota diagrams for VVT operation, you'll notice the range of cam movement allows the stock cam to be retarded to the point where it opens several degrees after TDC. So a performance cam that doesn't go too crazy on overlap can likely maintain near stock combustion quality characteristics at low loads. I mean this for both intake and exhaust. We lose a little bit of scavenging like this, but maintain driveability and increase fuel economy.

Originally Posted by serialk11r (Post 394968)

Oh, the heads might actually be a problem on this engine. Direct injection is picky about how the air flows into the cylinder, so you may completely ruin basically everything below 3000rpm or so if you port and polish. arghx7 said that the ports are probably "medium flow" in the sense that they introduce some tumble motion but not as much as typical DI engines thanks to the D4S setup.

Right now I am guessing the power is limited primarily by cams (and intake manifold), but once you start increasing the rev limit the heads might become the bottleneck, and there's pretty much nothing you can do at that point. The next FA20 supposedly has tumble generator valves and no port injection to address the issue with tumble flow and DI not working well at low engine speed, but that no doubt sacrifices some flow as well. How much we'll have to see.

My guess is Subaru will try to pursue high end power in the next FA by putting high flow ports with TGVs for low end combustion stability. I think the next FA20 might be something to anticipate, for this reason (it also has cooled EGR by the way, so there is no benefit to using D4S if they pull this off correctly). TGV and high flow, sounds like an oxymoron, but this is the only explanation, they could just use high tumble ports instead if they didn't care about flow.

Originally Posted by moto-mike (Post 1405163)

Frankly I think past 200 on pump gas is definitely optimistic. We hit an honest 205 or so on E85 with Simmons' car last year (now Derek's) and I doubt you'll see too much higher. It is > 100hp/liter, to the wheels, which is damn impressive. While we could get more creative with some dyno work (such as arbitrary dynojet conversions on other dynos...) I think there's just a limit to what a 2.0l will do with stock compression--which is already a ludicrous (for an OEM) 12.5:1.

I doubt you'll see too much with cams; the stock setup is already pretty aggressive. If it weren't you wouldn't get the torque dip nor spin to nearly 8k without a massive tq drop. Thankfully the dual VVT keeps things at bay. I'd say no more than 10 hp from cams alone. Compression wise there isn't too much room if you plan on doing many miles.

ITBs sound nice, and make for great response. But the stock plenum ain't half bad for flow. Larger throttle body(ies) might make more HP but I've not seen anyone prove this just yet.

My gut tells me Subaru/Toyota are not making this one easy for the NA aftermarket. Though they certainly did for FI, given the punishment these motors are taking.

Originally Posted by serialk11r (Post 396944)

Where did you hear about ITBs being stock? That sounds like nonsense.

You can't say 86mm stroke is perfect for high rev, but it can be brought to 8000+ if everything else allows.

It appears we have a rod:stroke ratio of about 1.51-1.52, which is close to the smallest number you'll find on production engines. The actual difference between the piston motion at 1.51 and 1.7 isn't that big mathematically speaking, but OEMs refuse to let an engine with a short rod rev, something to think about.

Originally Posted by Matt@Cosworth (Post 2491891)

looks like the bigger valves are the key to their 9,000rpm builds that plus a bunch of undisclosed oil system mods

high revs are always nice though......

Originally Posted by Toyota86.ir (Post 2492342)

It seems the Head is the limit.
i want to be sure before project.
i need to see dyno result

Originally Posted by celek (Post 2495236)

Head is not the limiting factor it flows around 270CFM at .470 Lift in stock form the head flows fine and with minor work it can make 700+ WHP with a turbo. Bigger valves are a waste in an NA platform. The limiting factors are cams, springs retainers, rockers and the factory intake manifold.