High HP NA

[subytech]

Just noticed this was quite the hot topic in another post and it peaked my intrest. I know it can be done just wondering how everyone would do it, and what kind of horsepower would be possible. And please no general answers like anything is possible with the right amount of money i think we all understand that.

[livestrong14]

2.5 shortblock, DOHC heads. Lots of porting and headwork, cams, good valvetrain, bigger injectors, lightweight pulley, lightweight flywheel, PP6 or other EM, custom intake setup, 2.5 inch straight pipe exhaust.

[subytech]

So a phase 1 2.5 with lots of emphasis on the cylinder heads interesting, what kind of compression?

[livestrong14]

10.8-11.1 I would say!

[subytech]

sweet, I wonder what kind of reasonable RPM these motors could run, I love motorcycles my zx6r redlines at 13.5K and my RX-7 was awsome too, I've always wonderd what an high rpm subaru would be like

[livestrong14]

With a good valvetrain, (springs, retainers, etc.) You can easily get 7,500 RPM out of them.

[SemperGuard]

You would be better off sourcing a newer ej204, than an ej25d. Stock they make about 190 hp. They have AVCS so mods would be more helpful on that engine than an old style cam engine.

[evolutionmovement]

The smaller engine would be better for high rpm as well and if you just want hp the higher rpm you can breathe at, the more you have.

[bmxpunk]

2.5 shortblock, DOHC heads. Lots of porting and headwork, cams, good valvetrain, bigger injectors, lightweight pulley, lightweight flywheel, PP6 or other EM, custom intake setup, 2.5 inch straight pipe exhaust.

Ok we are in the middle of doing this right now. THe Phase 2 block is supposedly much better if you are wanting to make bigger power. That said our first build is with a phase 1 block. Use the phase 2 block and dual cam heads with stock phase2 pistons and your compression ratio is close to 11.5:1. ultimately why would you use the dual cam heads? again it is what we are using on one of our 2 NA builds but the dual cam and single cam heads flow...almost the exact same. Shims or solid lash adjustment...I would go single cam. NOt to mention that do you want to buy 4 cams or 2? Drop in a set of the spicy cam from CObb. have your head work done by someone who knows what they are doing as this is where you can gain or lose huge. Stock injectors should be good for a while just add a walbro. Matts similar setup without engine management + ported heads cobb cams, the Air/fuel ratio was very good and I think still on the rich side. ON our build we are going with the Greddy Emanage ultimate, as I had no luck the the PP6. Equal lenth headers are an important part to the equation.

DO a search for subie sports NA 2.5 build something went wrong with their build and it should have more power but it points you in the right direction.

So here is the first NA build we are doing with another (single cam) following in the next couple of weeks.
Ph. 1 short block stock
Phase 2 pistons w/ceramic and moly Coating (final CR of about 11.5:1)
Ph.1 Heads ported and polished w/ ceramic coated CHambers/ valves/ exhaust ports
Webb Stage 1 cams
Reworked ecu w/7200 Rev limit and Greddy Emanage ultimate
Walbro Fuel pump
Ported intake manifold powder coated red
OBX equal length header
Stromung dual tip 2.25 catless Exhaust


hope that Good luck with your build

[SemperGuard]

The smaller engine would be better for high rpm as well and if you just want hp the higher rpm you can breathe at, the more you have.

The new EJ204 makes peak power at 7100 rpms. Redline is 7500 rpms. Throw on lightweight parts, spin it to 8500-9000, and keep modding.

[evolutionmovement]

Does that block have vertical bracing on the cylinders? I wonder how much of an issue that would be at such high rpm. Maybe it would only rear its head after many miles, but with the valvetrain you'd need, you'd be in there for top end work with some frequency anyway.

[SemperGuard]

It's an open deck like almost all other subaru engines. I'd be more worried about the bearings than anything else at that speed.

[555BCTurbo]

Something that a lot of folks are overlooking is port and intake manifold design...


Because of the crappy intake manifold design that Subies have (i.e. lots of curves) it is difficult to get a nice direct, fast charge going into the intake ports.

I would really like to build an NA Subie motor with individual throttle bodies, with velocity stacks on top and do away with the stock intake manifold.


Also Devan...if you are wanting to rev something...the SOHC 2.5 heads would be the way to go...they are roller rocker/solid lifter...and could pretty much go until valve float occurs.

[SemperGuard]

If you're stuck on SOHC, look at EJ259 intake ports. If you're wanting to do it right, get AVCS heads like on an EJ204. Dual AVCS would be ideal. ITB's are really only needed for throttle response. More important would be things like resonance, frequency, and timing for intake and exhaust design.

[Matt Monson]

Do an internet search for the old Syms build. I don't remember the details of it but it was something like 280hp@9000rpms from. 2.0l with itbs.

I've been contemplating a new NA build of late. I've got an '89 ej20d that is supposed to be 150chp@ 6800 with a 7200 rpm redline. I'm interested to see what those heads and cams would do with 2.2l of displacement. Not likely to break any records but you could probably hit 200chp with that engine and bolt ons...

[legacymax]

personally, I would go with a 2.2 if I was going for high RPM because of the shorter stroke.
-lightweight crank,
-lightweight rods
-Lighter, high compression pistons.
-hi rpm cams
-stiffer valve springs
-P&P DOHC heads
-turbo oil pump
-oil cooler

Most your driveability goes out the window but I would bet it would rev to 9.5k
-Max

[SemperGuard]

I just thought about what I was suggesting. If you're looking for just a maximum number in a narrow range of the rpm, then you don't need AVCS. In fact it would probably be worse if all you were going for was that, as the cam sprockets are a lot heavier than even the regular metal ones, not to mention the plastic ones. I think a 1.8 or 2.0 with the lightest stuff you can get would make good power in the 9000 rpm range. The stroke is the same from 1.8-2.2 so you might be better off going with a smaller bore so you have less mass, and can spin faster easier. Plus if you're worried about it being open deck a 1.8 has way more cylinder wall thickness than a 2.0 or 2.2. So if you're just going for high number at a specific place don't worry with AVCS, or ITB or any of that fancy stuff. Get the cams, intake manifold and exhaust manifold all setup for that specific rpm.

If however you're looking at getting great driveability all around, then go with AVCS, AVLS if you can figure out how to make it, ITB, and all the other goodies.

[Matt Monson]

Semper,
I totally agree with your logic. A lot of the Honda racers have followed the same logic for years and use non-vtec cams on racecars. While Porsche's and Audi's variable cams and runners have true performance value, most all of the other manufacturers' products are really emissions and fuel economy components...

[legacymax]

Matt, you would not want to run SOHC heads at all right because the rocker arms are not ideal for HI RPM, correct? Solid bucket DOHC heads would be ideal, correct?
-Max

[Matt Monson]

Matt, you would not want to run SOHC heads at all right because the rocker arms are not ideal for HI RPM, correct? Solid bucket DOHC heads would be ideal, correct?
-Max

I dunno. I know that some of the more reputible tuners say that the SOHC Ej25 heads can handle 9000rpm with no problem. I got mine up to 8500rpm once. Funny things happen when you miss a 4th gear downshift and hit 2nd instead. Heads had less than 5 miles on them at the time and I thought I had just destroyed my brand new build. But there was no ill effect and I kep driving them after that with no issues.

But back to your point, I still don't really know. I think you are probably right about the DOHC solid buckets being the best. I have always thought that the most direct valve actuation was the best approach for high revs and keeping valvetrain weight down...

[206er]

as far as total valvetrain weight, the sohc is probably slightly lower but it does have more reciprocating mass ie rockers. if you look at really high revving motors like bikes and things those are using a dohc setup. of course, 9000rpm is nothing to scoff at.
what can 2.5 bottom ends rev to?

[bmxpunk]

What about company like xceleration (spelling?) who offer a 2.0 WRX block with sohc heads built for 10,000 RPM?


but i do agree with the logic that solid lifter/buckets sound like the best idea, but that s not easily doable as those heads have different mounting for the cam sensors. acvs... etc as far as putting them in an older car.

At least not without standalone em.

WHat about the old dual cam heads with say, TWE's shim under bucket conversion? I know the motorcyle I am trying to get is shim under bucket (yz250F, 13,500rpm!!)

[evolutionmovement]

Screw valve springs - go desmodromic.

[subytech]

Screw valve springs - go desmodromic.

Hell yea now were talking,

[evolutionmovement]

Hey, if money's no object...

But I'd do step-motor-controlled rotary valves if we're talking tons of money and not caring about curing cancer. Sleeve valves might also be interesting, but that would require an entire block redesign, so it would no longer really be a Subaru engine project.