Info on my turbo engine build up

[ciper]

Didnt write this ahead of time, Ive done tons of research so ask questions if I forget to mention anything.

In short my goal currently is
91 turbo legacy engine as a starting point
Replace all wearable items no matter condition
02 WRX heads, bored to 96.9mm ringed and shortened to reduce head volume
Metal head gasket, .8mm preferred but most likel 1.52mm with arp studs
4-2-1 headers, slightly larger turbo, 3 inch twin dump DP/CAT/Resonator/Muffler
~500CC injectors
Air water intercooler with temperature sensative two speed pump and large exchanger
Custom spark plug wires to match now center mounted spark plugs, diamond coil pack and MSD ignition module. Staging RPM spark limiter which "should" act as a anti-lag (still have to test this one)
6 speed manual transmission non RA (standard viscous center diff) with mechanical front diff and viscous rear or
Level ten built Automatic
Electronic boost and fueling management
Plenty of gauges

Small internal parts will be new stock items (rings/bearings and such).

So far one part I need help on is sourcing the head gasket. Here is my dillema

The engines I have seen, ej25 and ej22, have the pistons flush with the deck surface when at top dead center. I would assume the ej20 to be the same. This means we only need to calculate area of the new engine combo.

Look at these numbers, tell me what you think
EJ22t stock bore 96.9mm stroke 75mm head chamber? 40cc piston volume 28cc gasket thickness 1.52mm calculated compression 7.98:1 published 8.0:1

EJ20t stock bore 92.0mm stroke 75mm head chamber? 49cc piston volume? 17cc gasket thickness 0.80mm calculated compression 7.99:1 published 8:1


The head volume of the ej22t was a guess based on the published compression ratio. The 49cc head volume for the ej20 was published by cobb tuning. The piston volume was guessed based on published CR.
Obviously these numbers should be verified but it gives us an idea. Not I calculate my new engine to have:

EJ22 combo bore 96.9 stroke 75mm head chamber 49cc piston volume 28cc gasket thickness? .152 calculated 7.27:1 CR

Realize this also doesnt consider the rebore of the head from 92 to 96.9 which will reduce compression even farther, 7.22:1 depending on how "deep" into the head. This is lower than I want. If I was somehow to use the smaller .8mm gasket from an ej20 the compression would raise to 7.67. As you can see a small change in gasket size makes a "large" difference in CR.
My current solution involves shortening the height of the head, basically milling the entire head surface down. I have to look closely at the valve lift because I dont want to create an interfearance engine in the process!


At first I was also planning to create my own stroker kit. This is still a secondary option.
Stock EJ22T has a 9.69 bore and 7.5 stroke

Stock EJ25 has a 9.96 bore and 7.9 stroke

If we kept the EJ22t pistons and used the EJ25 low end we would have:

Displacement = Stroke * * (Bore / 2)^2 * [# of cylinders]

Displacement = 7.9 * PI * (9.69 / 2)^2 * 4

Displacement = 7.9 * PI * 23.474025 * 4

Displacement = 7.9 * PI * 93.8961

Displacement = 2330.36805388979 CC

2330.36805388979 CC = 2.33 Liter (rounded)

2.33 liter is a small increase, but it seems that it would be a VERY cheap increase.

2.33 liters plus better flowing DOHC heads compared to the stock EJ22 or the increased Bore of 4.9mm + stroke of .4mm compared to an EJ20 is significant.

This power will be put to the ground with some 215 45 r16 falkens, compared to the stock 185 70 r14 thats 15mm reduced diameter or in other words a nice mechanical advantage

A worldwide switch replacement at work is has killed much of my time. One office alone has 9 Cisco 4507 with 288 10/100/1000 ports each (yeah gigabit to the desktop) and two 6513 plus a random assortment of 3500 smaller switches. I also just moved to the other side of the bay. My XT6 broke one of the timing belts (it has two) and my other legacy needed a new alternator and battery. Ive also been helping my friend replace basically the entire front drivetrain of his MX6. The BBS was something I just didnt have time for even though I had some questions to ask. One funny problem I had was my other Legacy's sunroof got stuck open! After taking alot of stuff apart I found out a light punch to the middle of the ceiling makes it work perfect

The "joint venture" GL-10 turbo 5 speed my friend and I had has been donated. Too many crazy full throttle runs made this tired car ready for the graveyard. It still ran when we donated it, upside down WRX muffler and all

We decided to weld front and rear differentials of a legacy wagon we got (traded for a pentium II laptop) and use it as our new dirt thrasher. We also have a puke smell blue wagon with a recently blown headgasket that will probably be stripped. For some reason we cant get rid of the puke smell!


This is probably a tenth of the information Ill share. The custom gauges and electronics Ill be using will be left for another post.

[legacy92ej22t]

He lives! Nice to hear from you ciper! It sounds like you've been very busy. You're car sounds like it is going to be a monster Nice. The 6-speed sounds killer. 2.33 litre, yes. I like what I hear.

[THAWA]

Yes very cool project you've got goin on there, sucks about all the extra stuff getting in the way though. Just curious, are you having to do anything special to get the car registered/smogged/whatever else you need to do?

[morgie]

After taking alot of stuff apart I found out a light punch to the middle of the ceiling makes it work perfect

yeah mine's worked on punchs also (the only way to get it work) .. there is a crappy connector located just over the driver-side sun-block / shield...

another solution : in the rear ceiling, in the middle of the roof, there is a plastic plug. Remove this, and with a hallen key, you can open or close the sunroof manually


Nice project ! keep us informed about that head swap, since i'm still considering it in my own project !

[vrg3]

My sunroof sometimes won't open, and the way to fix that is to push down on it from above. :)

[Matt Monson]

ciper,
Thanks for the info. I will be eagerly listening as this plays out since it isn't too different for my plans with my Turbo Legacy once the RS gets fully built...

My one question to throw out there at the risk of tread jacking...I will be using an EJ257 shortblock in my RS and will have the pistons left over. Would those be swappable into the EJ22-T for a better than stock quality piston? Have you considered something like that instead of aftermarket pistons or are you replacing with new stock ones?

[ciper]

Matt Monson: I have no plan to replace the stock pistons. Ive heard of pistons from other engines and even aftermarket units getting damaged but I cant remember anyone who had an EJ22T stock piston brake.
In the stroker kit I want to get some better rods, but keep the stock pistons and crank (ej22t piston, ej25 crank).


THAWA: To start the car will be with stock components (except for engine internals) and I will do an Engine Change through the DMV. Once inspected The exhaust and intercooler will be added. Then boost control and tuning electronics.
The part that is taking the longest is deciding on how to build the engine. Its already torn down and on a stand as we speak.

[Matt Monson]

I thought the crank and rods were forged and the pistons were a lower quality cast piece. Am I misinformed?
Of course for your stroker, you would have to which the crank, however...

[THAWA]

Coolness but what are you going to do when you have to smog it again? remove the extra stuff or do you have someone that'll look the other way on visual inspection?

[scottzg]

That is an incredible project and a lot of research!!!
I don't understand something though-- to what ends are you building?

A 6 speed tranny is nice, but expensive. Therefore you are not all that concerned with the budget.

You aren't building a drag racer because you are interested in a medium sized turbo.

You aren't building a autox'er/backroads monster/open wheel racer because you are putting in a awic, and they lose efficiency as they are cycled.

Based on your interest in mid sized turboes, awic, increased displacement, etc, i'd guess you were building an engine that approximated a big v6. Why? Subarus are not that good for midrange.

It just seems like a big expensive excersize in engine design.

I hope im not comming across jerk-y, im curious.

[entirelyturbo]

Phew! Talk about some research! Had to re-read over some of those numbers twice myself!

Ciper, I hope you're going to really be pushing the boost to this thing... If you've only got ~7.2:1 compression, your bottom end will be crap, and that miniscule displacement increase won't help much... Apparently you definitely aren't building this as a bad-ass daily... And the 6-speed is basically our STi's tranny right? Or not? You didn't mention anything about DCCD...

Also, your headwork... I was asking Morgie about it, as he's looking into EJ20T heads on his EJ22T. He said the piston is slightly above the block at TDC, so you had better choose your headgaskets wisely, otherwise you could have the piston slapping your head!!! As far as interference is concerned, however, the EJ20T heads, and any Subaru DOHC heads are valve-valve interference, so a broken timing belt would cause untold amounts of damage...

But, Scott, I really think Ciper is just screwing around to satisfy your fascination, and he's making a beast, who am I to blame him?

[Larry Witherspoon]

Can't remember the details but with overhead cams, a milled head can cause a change in valve train gear position relative to the crank and other driven accessories. A little bit off and you might get some problems, related to valve/cam timing and belt tension if not compensated, or something like that.

I'll keep my eyes peeled and if it's serious I'll get back to you.

What sticks out in my mind was the statement that milled heads on an overhead cam engine should not be taken lightly.

[morgie]

A friend here had a 2.5 RS 98 with the DOHC heads . He got the heads milled onces, and that event started is nigthmare... Problems, problems, problems... blown head gaskets, air in water... etc ... got the heads milled 3 times.. :\

Finally, his nightmare ended this spring when he installed a EJ20T into his car.

that's a big reason why i'll never do that on my engine if i decide to put DOHC heads on my ej22T.

[ciper]

THAWA: Smogging it again will be tricky but not hard. The car will have a valid engine swap certification, the intercooler will be subaru and the engine bay will look normal (I dont beleive in yellow hoses and painted intakes BLEH). The car will have a cat+resonator+muffler equal to the original number. The electronics will be out of site as will all the gauges except for EGT which is very inconspicious, especially since Im using an aviation unit (no fancy ricer parts). I plan to have the emissions stickers handy and the car will EASILY pass the sound requirements. The visual inspection should pass since all the emission controls will be active. I even plan to keep the stock intake box with panel filter. The MSD ignition has a Carb excemption. The headers are what may mess me up.

scottzg: Im building it to enjoy. The transmission choice is later, an older 6 speed without DCD (standard viscous center) is cheaper than you think and a built up automatic isnt far behind. I already have all the other drivetrain components, include spair ring and pinion sets of 3.7 3.9 4.11 and 4.44 with matching open or Viscous units and axle stubs.

I dont plan to compete in any fashion, heck even when I rallyX or autoX the car I was in the open class but I didnt care. I was doing it for fun.
The v6 comparison makes sense, id take it more like a 4.6 v8. I want good torque across the range. I dont mind the work in designing the engine as long as it works when Im done. I dont plan to follow the route already taken by others, but rather combine the knowledge they have gained.

I think you came across very well, I apreciate someone you can get to the point!!


subyluvr2212: I dont plan to keep 7.2:1 compression. In the end I hope for 8:1. If need be Ill get shallow dish pistons.
The "small" increase in displacement will make a BIG difference.
Two ways could be used. If I increased the bore I would agree its not worth it, in the end it would be a waste of time. However Im gaining this displacement in STROKE. Meaning very long exhaust pulses that will spool the turbo alot faster.

Larry Witherspoon: Thanks for your input, that is something I had not thought about and will research. I planned to remove 1-2mm. Please check on it for me.

morgie: I wonder why it happened. The heads probably didnt have much damage so they wouldnt have had to mill very much. If the block mating surface had no damage then I cant see why it would cause the gasket to leak worse, unless he reused old head bolts or something. You think the head bolts may have bottomed out causing a torque reading that wasnt accurate?

[entirelyturbo]

THAWA: Smogging it again will be tricky but not hard. The car will have a valid engine swap certification, the intercooler will be subaru and the engine bay will look normal (I dont beleive in yellow hoses and painted intakes BLEH). The car will have a cat+resonator+muffler equal to the original number. The electronics will be out of site as will all the gauges except for EGT which is very inconspicious, especially since Im using an aviation unit (no fancy ricer parts). I plan to have the emissions stickers handy and the car will EASILY pass the sound requirements. The visual inspection should pass since all the emission controls will be active. I even plan to keep the stock intake box with panel filter. The MSD ignition has a Carb excemption. The headers are what may mess me up.

This is exactly what I would do with something like this! Make it look as professional as possible, like Subaru did it themselves. The one thing I hate most in modding cars is half-assing it. That's the main reason why I don't mod my car much. I don't really have the money or time to REALLY do it right... That should change within the next few years though

subyluvr2212: I dont plan to keep 7.2:1 compression. In the end I hope for 8:1. If need be Ill get shallow dish pistons.
The "small" increase in displacement will make a BIG difference.
Two ways could be used. If I increased the bore I would agree its not worth it, in the end it would be a waste of time. However Im gaining this displacement in STROKE. Meaning very long exhaust pulses that will spool the turbo alot faster.

Yeah I think that would be an ideal CR... just enough torque to move the car along off-boost... I also see your point increasing the stroke instead of the bore... Good idea

Just curious, ciper, exactly what engine management are you using? Piggy back, stock ECU reprogram? Or you can keep your secrets to yourself if you like

[scottzg]

That's very cool- its probably the smallest way to create an engine with a big round power curve. I wanna ride when you're done! How are you going to set up the suspension?

That doesn't mean i don't think you're crazy. There are lots of cars with big fat power curves directly out of the factory (and a lot of them nicer than 10 year old subarus)- the implied mustang GT, for instance. Sure is a lot less headache.

I guess it's funny to me that your knowlege is so extensive and your ambitions are so modest.

[Matt Monson]

Like Ciper said, getting that car by CA emissions shouldn't be that hard. The BAR referee will have no way to tell that the internals and the heads are not stock. If the number of cats is right and all other emissions are in place, it shouldn't really throw any red flags.

The biggest concern would be the engine management. If this car is going to have a stand alone ECU like a Link or something running it, it should be very well hidden because the ECU will fail you.

A little trick we used to do when I lived in the Bay and did Honda engine swaps was to present a Civic that was swapped, but was running bone stock. After you got the BAR sticker on the doorjam stating that your B16 was legal, then you could do the headers, piggy back electronices and whatever else you wanted because no smog tech at an emissions station would ever question it with that little BAR sticker on the door jamb. All they care about is whether or not it passes the sniffer. That might be something to consider on the headers. Switching headers is only an hour or two operation. Also read this before you spend a grand or more on some fancy JDM headers...
http://www.cobbtuning.com/tech/exhaustdesign/index.html

My build isn't going to be that different from this, but I am sticking with the stock headers. I go back and forth on SOHC EJ25 heads or DOHC EJ20 heads. They both have benefits and liabilities to them. Personally, I dream of a 5EAT tranny in my wagon with the paddle shifters on the steering wheel. And like Ciper, this will be my daily driver that just has character,uniqueness and of course is totally sleeper. My race car is gonna be the RS...

[ciper]

scottzg: The rest of the car is basically done. Name a part with an aftermarket equivalent and I probably have it already. After this comes getting the dents removed and paint touched up.

I actually plan to own a 5.0 miata one day which will have the big power curve and great weight to match

Matt Monson: Exactly.
I dont plan to get any fancy headers, they will be the smallest diameter units with a closer to equal length as possible that are compatible with the oil cooler.

Why not DOHC EJ25 heads?

I think I want to do the EJ25 heads but I need help in sourcing the tighter valve springs. Anyone know of a source?

[Matt Monson]

http://www.techworkseng.com/news.html
These guys up in Calgary sell springs.
As for the DOHC question, there is no reason not to go that route. This thread here:
http://www.wrxforum.com/cgi-bin/ultimat ... 000917;p=1
and
http://forums.nasioc.com/forums/showthr ... did=448397
Will give you some food for thought. I am going SOHC for the reasons mentioned in those threads and the fact that it will save me some $$$ since I will have a spare SOHC EJ25 head on my hands once the RS is built!

[Legacy777]

not sure if you've read larry's info about the ej25 head swap....if not, it's a very informative read.

there's some various threads out there.

http://bbs.legacycentral.org/viewtopic. ... ej25+heads
http://bbs.legacycentral.org/viewtopic. ... ej25+heads
http://bbs.legacycentral.org/viewtopic. ... ej25+heads
http://bbs.legacycentral.org/viewtopic. ... ej25+heads

there's some other threads as well, but you can search in the archives yourself for them.

[ciper]

Matt Monson: Dammit, why do you have to through yet another question into my head Now I have to research those as well

[Matt Monson]

Matt Monson: Dammit, why do you have to through yet another question into my head Now I have to research those as well

What can I say...I myself overthink everything before I actually do it. It took me almost two years to settle on what I was doing with the RS. And still couldn't commit completely to an NA build so bought a Turbo Legacy to build too...

[ciper]

Okay, the EJ25 stroker kit is now something I dont want to do. The strength of EJ25 rods and crank are questionable when compared to the EJ22T items. It looks like custom pistons will be needed either way so Id rather not get a custom Crank and Rods and Pistons!! The price for performance calculation shows that my money would be better spent on the pistons and head modifications if I had extra.

Since I will be getting custom pistons to get the correct C/R increasing the bore is easy. I now have to figure out how much material can be removed and if any sizes will be cheaper. For example if I was to use the stock bore of the EJ25 I would have:

9.96 EJ25 bore 7.5 * pi * (9.96/2)^2 * 4 = 2337.38 displacement
As a comparison, an even number
10cm Bore 7.5 * pi * (10/2)^2 * 4 = 2356.19

Thanks to Matt Monson and the information I found SOHC heads are now the current route.

So the short recipe is:
EJ22 block with stock internals, wear items replaced (bearings and such) bored
SOHC heads, possible cam replacement for more turbo friendly overlap, possible tightend springs
Custom pistons to match bore and head volume (C/R)

If I use SOHC heads it will look more stock than ever, plus I can use normal spark plug wires and timing belts!

[Matt Monson]

Your welcome...now to plant another seed of contemplation...I am using EJ257 pistons( I will have them spare after my RS build) in mine. I am not worried about the slightly low CR for reasons mentioned in some of the other threads on this subject. But I also know of a set of JE 8.5:1 CR 100mm pistons for sale below cost...more math...

post 100, bu who's counting, right?

[ciper]

Matt Monsoon: EJ257, meaning usdm 2004 2.5 turbo STI engine? Sexy
You say slightly low compression, do you have the specs on the STI engine and yours to calculate what it should be at right now?


JE 8.5:1 CR 100mm pistons, that 8.5:1 is based on if they are used in what exactly? My inexperience in the Subaru scene is showing

Im looking to keep C/R above 8.0 especialyl if I can get the quench clearances very low.

BTW read this excert from a TWE email I got. These guys are smart!

For head gaskets, we have been using a combination of the DOHC and SOHC gaskets. We add or remove the stainless steel shims until “optimal” quench clearances are achieved. Optimal meaning just close enough that the piston doesn’t hit the head. Optimal for our full race engines means running the clearance tight enough to remove the carbon on the piston tops!