DOHC vs SOHC PnP w/reground cams

[206er]

okay I THINK my math is ok...

for the SOHC heads on ej22t with ej22t gasket,
total volume at BDC= 553cc(ej22t CC)+11.2(ej22t HG)+28(ej22t piston)+50cc(ej25SOHC combustion chamber)
total volume=642.2
total volume at TDC=642.2-553=89.2
compression ratio=642.2/89.2=7.1:1

for ej22e/SOHC
BDC:553+11.2+14.5+50=628.7
TDC:75.7
CR:628.7/75.7= 8.3:1 WOO HOO!!!

is this combo going to last? the open deck is probably a non issue, what about the pistons? I want to build one of these out of my old ej22e and rig up a T3/T4 and run megasquirt.

what are the:
SOHC injector options?
SOHC turbo oiling options?
SOHC manifold options?

[555BCTurbo]

The injector options are endless...as they use topfeeds iirc...same as WRX, JDM STi, etc

Manifolds are pretty much limited to the 2.5 and USDM WRX 2 liter iirc

Oiling the turbo would be the most difficult issue...but if you were to rig a line up from one of the outlets (i.e. oil pressure sender, etc) you could probably make that work

You also may be able to take the heads to a machine shop and have an NPT fitting placed in one of the oil galleys in the right head, and then just use a turbo oilpan with the drainback hose...or weld a bung on your stock one.

[sammydafish]

Oiling the turbo would be the most difficult issue...but if you were to rig a line up from one of the outlets (i.e. oil pressure sender, etc) you could probably make that work

You also may be able to take the heads to a machine shop and have an NPT fitting placed in one of the oil galleys in the right head, and then just use a turbo oilpan with the drainback hose...or weld a bung on your stock one.

that's exactly how I oil the turbo on my NA-T scoobie. Oil from the sending unit and I welded a pipe that runs to the oil pan for return (by the way, this is pretty tough as there's not a lot of room to do this. I have solid welded motor mounts, so I cut about an inch or so off of one to make room

[Matt Monson]

Damn,
So much content to address in this thread since I last replied. I will start with a picture:


That's the oil return for my Rimmer Supercharger into a SOHC head. You could do the same for a turbo. My oil feed comes off a splitter on the pressure sensor on the top of the block under the steering pump.

Now for the real issue with using SOHC Ej25 heads: Up-pipe. A stock uppipe when using a turbo crossmember, clips the back corner of the SOHC heads. I know of one guy who has cut out a chunk of the head to make it work, but everyone else I know that has done this is using custom piping to make it clear. I don't know how the twin scroll parts would clear.

206er,
Your math is dead on. I know of a handful of guys using the Ej22E with Ej25 heads (the DOHC's are 8.6:1 because their combustion chamber is a little smaller). They run 14-16 psi on them and they are mean little cars. I was gonna do one, and then had to use my Ej22E bottom end for emergency repairs on a buddy's car.

To clarify my earlier comments that have raised some questions...I am saying that in stock form the SOHC Ej25 heads will EASILY make the power level you are looking for.
http://www.cobbtuning.com/tech/sohc/page2.html
Here's Cobb's bench flow with both the EJ22T and SOHC heads. You can see that the EJ25 heads outflow the Ej22T heads by almost 50cfm on intake and 35 cfm on exhaust. For reference, my ported SOHC Ej25 heads make 272 cfm intake and 216 exhaust. That's +/- a 20% gain. Assuming you can get the same gains out of ported Ej22t heads, they will come up to roughly the same flow numbers as the stock SOHC Ej25 heads. It's really a crap shoot which route you choose to go.

I do NOT favour porting ANY of the EJ25 heads for this project. Xephyr, and other friends of mine broke 400whp before they ever went with ported heads. I stand by my opinion that it's all about the turbo. And unless you have a HUGE turbo, power will fall off above 7000rpm pretty damned quickly.

To use the SOHC heads, will require any post '99 intake manifold. The WRX/STi bolt pattern is the same. This makes the injector options pretty limitless. The injector option for the SOHC manifold are pretty limited UNLESS you send your manifold out to someone to machine it to fit the other top feeds out there. For my SC project, I am using a JDM v6 STI manifold (w/SOHC heads) because I can drop in the yellow tops very easily. If one were to use a WRX or STI manifold, not only would you be able to change injectors, but you plumb the intake under the manifold and not deal with the whole 90 degree bend issue we struggle with around here.

I am sure there is more that I have missed, but need to go back through the thread and read again and respond to other holes, even if just to confirm things others have said...

[93forestpearl]

Matt, how limiting will my choice to stick with the stock heads be, but with the cams reground by Delta? I wasn't planning on updating my heads untill I went for crazy power, above the 400bhp or so that my GT2871R .86 A/R is capable of.

I don't know enough about porting to just dive in with the die grinder, and I don't know of anyone up hear that would know what they are doing with these heads.

At this point, updating the heads isn't really an option for me, since several of my choices have revolved around using the stock heads with injectors that fit. Also, I'd like to think my choice of turbo, rotated mount, and no MAF sensor will help with my current goals.

[Matt Monson]

I know Colorado locals who broke the 300whp barrier using unported stock EJ22T heads.

If you are willing to strip your heads and ship them off to me as just bare cores, I can get them ported for you for less than $600. If you wanted a valve job on top of that, roughly another $200, and you'd need to include all the valves...

[BAC5.2]

So Matt,

I think I misunderstood you earlier.

You are more a fan of using stock-ish heads and the "right" turbo and just forcing air through the ports?

We had one WRX with an EJ257 bottom end and EJ20 heads with an FP18G make 350whp and 370lb-ft. It made all of the torque at 3100RPM. With the stock EJ20, he made 330whp and 270lb-ft on the same turbo. He jumped 100lb-ft at exactly the same RPM. ALL was done through a top mount intercooler.

Is it your belief that any power number is simply a turbo choice rather than a head working choice? Obviously this has a limit, and something outrageous like 700whp won't be easily acheived using stock heads and cams, but for numbers up to and including 400whp, is that what your belief is?

That's cool. Thanks to you, I've seen both sides of the story, and where I stand is slightly leaning towards the built-heads and cams side.

Ultimately, I think it's a personal choice based on just what you want the car to be. SOHC heads flowing the same CFM as DOHC heads, I'd venture to guess that their performances would be almost identical to eachother assuming similar cam specs.

If you intend to leave stock cams in the heads, it won't matter. I think DOHC heads give you more versatility when it comes to cam choices.

With cams, does it stand to reason that there is less restriction from air entering and exiting the combustion chamber (using more lift and duration, the valve opens farther for a longer period, allowing more air to enter the combustion chamber)? But with cams, spool times tend to increase slightly (or dramatically based on specs).

Perhaps without cams, the increased restriction increases exhaust velocity slightly and that helps more efficiently spool the turbo, resulting in higher boost at lower RPM with a trade off of up-top breathing. If you never intend to break 7000RPM, this could actually be a huge benefit, as you see virtually no down side.

But if you want to hit 8500RPM, you will run into the problem of not being able to flow enough at higher RPM, and the addition of cams will help that scenario. But a turbo capable of performing at such a high RPM will take a lot of exhaust velocity to get spinning, and lower RPM performance will suffer.

Seems like a catch 22 to me. If you want 600whp and 8500RPM to play with, you are in a bit of a pickle when it comes to having some kind of low-end torque. A turbo big enough to produce that kind of power and be efficient at that RPM will run into down-low breathing problems. With stock cams, you could use a smaller turbo, but you run into the problem of the heads not flowing enough to handle 8500RPM.

Is there a perfect combo for every power goal? If I wanted to make 600whp by 7000 RPM, is that even possible (without nitrous)? Or would I have to do valvetrain work, change cam specs, and bump the redline up to 8000? When I do that, I'll lose valuable bottom end though.

What about cams in stock ported heads? Again, catch 22. The cams will help high RPM breathing, but the porting will be counter productive to high RPM breathing. Would that situation even be worth doing? Would there be appreciable gains over stock? Would the stock porting totally invalidate the new valvetrain (built for higher RPM usage)?

I don't think it's a matter of SOHC vs. DOHC, but rather flow rates and power goals, as well as other goals set by the user.

Can 600whp be made by 7000 RPM and still retain some form of low end torque, while at the same time not running everything at 10/10ths? Can 600whp be done without beating the shit out of everything, and can it be done by 7000RPM and can it be done and still make some appreciable power at a low RPM?

Without Nitrous, or some other form of turbo-insta-spooler, I think it might not be possible.

[sammydafish]

The really important question to ask, now that the flow question is answered in rearguards to running an EJ22T head or the EJ25 SOHC head is which design is more resistant to detonation? Chamber shape is the key factor in detonation resistance. It seems the EJ25 head has a larger chamber volume, therefore compression would be lowered, but comparing equal compression ratios, what head would offer greater resistance to detonation?

I'd assume the latter design as, well, it's newer and really a more refined version of the prior head. That's just an assumption though. This should be the most important factor in determining which head to run, aside from flow.


As far as flow is concerned. Anything you can do to increase the volumetric efficiency of you engine will allow you to make more power. Even on a forced induction engine. Say your engine NA flows enough air to make 150hp. Your turbo which can flow tons of air, in a perfect* world at a PR of 2 should allow your engine to make 300hp. Great, it's capable of it. But, if your engine with good flowing heads and other work flows enough air to make 180hp, than in order to make 300hp you only need to put out a PR of 1.67, or a LOT less boost. Less boost means less heat which means a more efficient burn and less detonation. OR look at it this way. Less boost for hp X means you can get a higher amount of power for the same boost/hear/detonation risk as you had earlier.

The bottom line is, you could make the power with less flow, but more flow will never hurt.

BTW, if you loose low end power due to greater flow this can always be corrected with cam profile changes. Remember, the cam was designed for the head as it was built, if you change it, it might not work the same.

[93forestpearl]

My heads are pretty much finished, I'm just waiting on new exhaust valves to get down to the machine shop. They did a 3-angle on the intake valves. I think I'll be fine with what I have. I don't want to put too much into these heads since when I get out of school, I plan on getting a bigger snail, injectors and updated heads. I was just worried these heads would severly limit my turbo's ability.

[Splinter]

Sammy, some really god thoughts there.


Im thinking my best option is going to be to get the 2.2L heads ported for better flow, and use reground cams to achieve the flow characteristics I need for that low end torque.

Ill run high compression and low boost, 7000rpm is a fine ceiling for me.


The reason I want to stick with the stock EJ22T heads, rather than the RS heads is becase of the plumbing issues. My power goals arent anything majorly crazy.

So to recap, the current plan is as follows:

Modified stock heads/cams
High compression pistons
650cc injectors
Front-mount intercooler
VF39
2.75" high-flow cat exhaust
Autronics ECU + MSD MIS-2 ignition


With proper tuning, that should be able to hit 300ft-lbs at fairly low RPM, yes?

[Matt Monson]

Bac5.2,
You cover a lot of information there and I think it's really good stuff. To answer your question is really a subjective one. I will once again make the example of Xephyr's car and his evolution, not just because he's a friend of mine, but because his thinking and approach is very similar to mine. His '98 RS was built to it's current form for 2 things. #1 as his daily driver. And with a full dog box and his power levels, it is one of the less streetable daily driver's out there. #2 He's a road racer who does time trials and time attack type stuff, and used to do a lot of auto-x until our local region got overcrowded.. As such, total power under the curve and good low end are really important. Now, if he were a drag racer, the "nature" of powerband that he was looking for would be totally different. That kind of setup would be peaky and and not worth a damned below 4500rpm, and likely make more HP than the other approach, even if the same turbo was used.

So, all other things being equal, I will describe the evolution of Mark's build, even though I think it may be redundant to what's in the sticky. Through the initial phases, his turbo remained unchanged, and he was using a Garrett custom turbo on unpublished specs. That is the one challenge when talking about Mark's setup; he is a beta-test monkey for TEC in Golden and none of his trim number or compressor housing, etc are released until TEC has decided to put that particular unit to market. And since what I am going to describe starts roughly 3 years ago, I am too lazy to go searching for the info on his exact turbo.


Set up #1: Ej22T bottom end w/stock DOHC Ej25 heads (which flow really closely to the same as the SOHC heads). For discussion purposes, let just say this is the set up he had 300whp at this point. The car had a nice fat torque band and solid power from around 3000rpm to his 7000rpm shiftpoint.

Set up #2: After a while, he threw in some PnP'd heads w/Web reground cams. He picked up something like 20 peak HP, and his peak moved up around 500rpm or so. But he lost a lot of low end and really the powerband became more like 4000-7500rpm. Narrower and more peaky. He was "forced" to retire the set up because the cam spit a lobe and destroyed his heads!

Set up #3: Back to stock heads, still with the Ej22T bottom end. This is when he upgraded turbos. This new turbo still allowed him that power down in the 3000rpm range, but suddenly he could rev up to 7500rpms and still make good power up there. He was making maybe another 20-25 peak WHP, with more usable power across the board. He really liked this set up and left his car unchanged for over a year, which is a really long time for Mark. During this time, he rebuilt his dogbox with Modena gears that were specific to his engine/turbo set up.

Fast forward a little bit. About 18 months ago, Mark decided that he wanted 400whp. I think his highest number with the Ej22T was around 360whp. He ran the numbers (he's an engineer and is meticulous with this stuff) and decided that physically there was just not enough potential energy in 91 octane pump gas to hit his goal with only 2.2l of displacement. I should mention that he has always tuned his car to run pump gas, and that was part of his goal. No point in a daily driver that needs race juice all the time.

Set up #4: He switches in an Ej257 bottom end. Still on stock DOHC Ej25 heads with nothing upgraded but springs. This set up puts him up to 381whp. He runs like this for a while, but knows that with the greater displacement a new turbo will be required in short order as he is needing to turn up his boost an extra 2-3psi above 6500 rpm because the turbo is out of it's efficiency range. Without that tweak, it falls off terribly and won't hold power to 7000rpm.

Set up #5: He throws in the some HKS v3 STI 272 degree cams. Same experience as last time, except his gains were even less than on the EJ22T set up. He picked up about 10 peak HP roughly 700rpm higher in his powerband and his low end just plain sucks. He disliked them so much, they were in the car less than 72 hours!

Set up #6 Back to the stock heads and cams again. He's now got a new turbo from TEC that is even larger. The specs are as follows:
Turbo: Garret 60-1 custom turbo built by TEC in Golden, CO
Compressor: late model .50 a/r, 60 trim 2.3"/3"
Turbine: 0.82 a/r, 76 trim, 2.2"/2.6"
This shot him over the 400whp mark to 452whp. I posted a thread in the ashtray about this in April of '05.

Set up #7 He got caught up in a little max. pump gas competition with some of the tuners on NASIOC and decided to give ported heads a try again. He got a good deal on a set of Cobb stage III turbo heads with race cams. These got him up over 480whp. But he hates it again. He says it isn't worth a damn below 4000rpms, and even though it will rev to 7500rpm, it just annoys him to no end. I would imagine that these parts will be off the car by summer...

I know this is long and drawn out, but just wanted you to see how much time and effort he has actually put into exploring it. It really does come down to are you looking for a peak HP number or trying to create a particular power curve? And it supports my idea that the neighborhood of the actual number is really dependent on your turbo, and then you can tweak it to your application from there for the type of powercurve you want. And the other thing that influences me is $$$. I figure why spend $1500 on headwork and another couple hundred on cams, when the change in max HP is likely only another 20-25hp. Why not start with a bigger turbo to start with, and save $1000 or more on the build and still hit the power goal...

[Splinter]

Oh its also worth mentioning that Ill be tuning to run on 94 octane

[evolutionmovement]

272 duration cams seems like too much overlap for a turbo car to me. Maybe the issue is the duration in that case, as only at high rpm would would that be beneficial at all and would really kill the bottom end with boost blowing partially out the exhaust. Higher lift might be beneficial all around, but then you're going to need some real beef top end components for high rpm.

Has anyone played with the variable valve timing on the STI? Is it a factory cam spec issue, complexity issue (there are plenty of stand alones to run variable valve timing), or durability? I'm not sure how the Subaru system works, but if it's like Honda's with two separate lobes, that could be the answer to the low/high end breathing compromise (which is the point of such systems to begin with). A timing only change wouldn't probably be worth the bother if that's the type of system, though (like my Mazda).

[Matt Monson]

With proper tuning, that should be able to hit 300ft-lbs at fairly low RPM, yes?

I think you will be able to get that number. I see guys running the Ej205 heads w/EJ257 swap consistently hitting 290-320 ft/lb on the dyno in the 3500-3800rpm range with that turbo. Once your Ej22T heads are ported, they should be flowing as well, if not better than the Ej205 heads. The Ej25 heads do outflow the EJ205 heads by a fair bit.

I am really interested to see this build take place. Nobody has really messed with the Ej22T heads very much in recent times. I am confident that they can flow up there with stock EJ205 and EJ25's. My only concern would be the HLA's are going to be an issue up there around 7000rpms, especially with the amount of boost you will be pushing. You will definitely want to do something about your valvesprings as well. Avoiding float is going to be your greatest challenge...

[BAC5.2]

TurboXS takes advantage of AVCS by using it like an EGR system, but better. Simply allow the exhaust valves to open slightly just before the intake valves open, and draw a little exhaust BACK into the cylinder. Helps prevent Det at high boost and high RPM.

Kind of neat, really.

Matt, I have a longer post in reply to yours that I'll handle tomorrow when I have more time. You bring up a lot of good information, and I'd really like to pick your brain a little so that I can attempt to fully understand what exactly I want to, and am trying to, do when all is said and done.

[Splinter]

So if Im running high, fast lifts with strong lifters, Ill have less headroom, RPM-wise?

[sammydafish]

Has anyone played with the variable valve timing on the STI? Is it a factory cam spec issue, complexity issue (there are plenty of stand alones to run variable valve timing), or durability? I'm not sure how the Subaru system works, but if it's like Honda's with two separate lobes, that could be the answer to the low/high end breathing compromise (which is the point of such systems to begin with). A timing only change wouldn't probably be worth the bother if that's the type of system, though (like my Mazda).

pretty sure it's variable cam timing, like VVT-i, vanos and others. Almost every manufacturer does this now and it's better for what we're talking about. Loosing low end torqe because of too much head flow and cam duration can be fixed by retarding the intake cam at lower RPMs (actualy it starts retarded and is just advanced at higher RPMs) The ability to do this for the most part eliminates the need for high duration cams. It alows you to have little to no overlap during low RPM and idle, yet tons of it at high RPMs. Best of both worlds. If the exhaust cam can also be controlled (most systems just turn the intake cam), then there are almost limitless possabilities for creating a near flatline torque curve and power that seems to never fall off. Its because of technology like this that cars like modern BMWs make as much power as they do with little displacement and no forced induction.

[evolutionmovement]

I thought the Toyota was a combo system like Honda's two cam profiles, yet with the ability to adjust timing within a given profile as well. That sounds best to rotary valves or solenoid actuated poppets

[Splinter]

I think you will be able to get that number. I see guys running the Ej205 heads w/EJ257 swap consistently hitting 290-320 ft/lb on the dyno in the 3500-3800rpm range with that turbo. Once your Ej22T heads are ported, they should be flowing as well, if not better than the Ej205 heads. The Ej25 heads do outflow the EJ205 heads by a fair bit.

Are you talking wheel or brake numbers? Adjusted or raw?

[sammydafish]

I thought the Toyota was a combo system like Honda's two cam profiles, yet with the ability to adjust timing within a given profile as well. That sounds best to rotary valves or solenoid actuated poppets

That's VVTL-i, (Variable Vavle Timing and Lift with inteligence) which was available on the Celica GTS. Not sure if they're running it on other models now. Dispite the sound concept, the gains for the added cost/complexity were minimal on the Celica.

[Matt Monson]

I think you will be able to get that number. I see guys running the Ej205 heads w/EJ257 swap consistently hitting 290-320 ft/lb on the dyno in the 3500-3800rpm range with that turbo. Once your Ej22T heads are ported, they should be flowing as well, if not better than the Ej205 heads. The Ej25 heads do outflow the EJ205 heads by a fair bit.

Are you talking wheel or brake numbers? Adjusted or raw?

I am talking about wheel. And those are torque numbers. The WHP is generally substantially lower like 250-260whp. Generally they are with a 1.15 correction factor. Obviously, with less displacment, you will see a bit of a drop in your numbers, but even 230whp is over 300 at the crank...

[BAC5.2]

Matt - One of our cars with the EJ257 bottom end and EJ205 heads made 344whp and 370lb-ft. Torque peak was around 3300 or something like that.

[Splinter]

I dont care about HP numbers
Torque makes cars move off the line, HP moves them off the car lot.

[BAC5.2]

Horsepower does indicate where torque starts to fall off significantly though. Two cars with the same redline that make the same 300lb-ft at 3000 RPM, and one makes 500hp at 5000RPM and the other makes 500hp at 6500RPM, the second car will accelerate harder, faster, and longer. Torque curve on the second one is drawn out longer and thus, when the first car's torque curve dies down, the second car is still on-boil.

Horsepower is a good indicator of just how quickly the car can effectively do the work. Diesels are good examples. 1000lb-ft and 400hp. They can be quick, but it's just ALLLLL torque, and not a whole lotta eagerness to put it to the ground.

Horsepower and the RPM it's at it's peak, really just helps give 2 dimentional structure to a plot in your head. High horsepower numbers at high RPM, in conjunction with high torque numbers at low RPM can tell you a LOT about a cars behavior.

[Matt Monson]

Matt - One of our cars with the EJ257 bottom end and EJ205 heads made 344whp and 370lb-ft. Torque peak was around 3300 or something like that.

With a VF39? What kind of dyno? what kind of correction? etc. etc. I was just giving him a reasonable expectation of what he would get out of it. There are always going to be outlier and extremes...