legacycentral bbs

Just wondering if anyone has tried or did put a turbo on a N/A EJ22? Not with changing the wireing harness or anything, basicly just bolting up all the turbo stuff and running it like @ 5psi?

The NA to turbo legacy swap has got to be the easiest swap ever, compared to any other vehicle. Only three wires on the engine harness need changed.

So are you saying everyone whos tried has just added the three wires to the harness? I didn't really want to add any other sensors for the purpose. Is this possible?

I actually thought the swap would be harder, I purchased a complete turbo car for parts to make sure I had everything.

I made a mistake, it needs two wires swapped (one for one cross connect) and two sensors added. The additional sensor wires are for the wastegate and pressure sensor.

Im having someone do the swap for me. Im capable but Id rather let someone else do the dirty work, so that I can the fun task of tuning.
Did you see this?
http://www.graphics.cornell.edu/~v/ecupins/

I think there may be some confusion here. One of you I do believe is talking about "slapping on" a turbo to a NA engine. The other is talking about swapping in an EJ22T. I may be wrong. But just in case, turbocharging a NA engine is not good. Finding a turbocharged Legacy engine and dropping that in, waaaay better. For many reasons too numerous to list again.

This has been discussed before.

I always think of this equation when asked this question:

N/A engine + turbo = Big Trouble $$$


It's just not a good idea of adding a turbo to an engine that wasn't designed for one. Even running the engine at lower boost still puts extra stress on the engine. If you rebuilt the N/A engine so that it can handle the extra stresses of the turbo, then you can do it.

Now hold on a second...

It's certainly true that an engine built for turbocharged duty (like the EJ22T) will be much sturdier, but people turbocharge engines that were naturally aspirated when stock (like the N/A EJ22) all the time.

My brother's in the process of turbocharging the 6MGE that he has in his MkII Supra right now; many others have done the same with that engine or similar. There are plenty of people on NASIOC with aftermarket-turbocharged Impreza 2.5RSes. And just glance through any import tuning magazine and you can see how popular turbo kits are for the wide variety of engines that come atmospheric from the factory.

You certainly can't push nearly as much boost -- 5 or 6 psi, or maybe 7, is probably about as high as I'd go on an N/A EJ22 -- and you have to be much more careful with engine management, but if you're careful about it you can definitely turbocharge your car. You have to remember that the engine is much more sensitive to knock, has a higher compression ratio, has a smaller cooling system, and is just weaker than the EJ22T.

Most of the problems that people have with aftermarket turbos are due to one of two things: 1) not doing the job completely, resulting in bad things like detonation, or 2) not recognizing that you have to treat the car differently.

But a properly done aftermarket turbo install won't destroy the engine. Of course it will increase wear a little and decrease fuel economy a little, but it can still outlast the body of the car. For a turboed EJ22 I would say an intercooler is a necessity. You also have to find a way to richen the fuel mixtures when on boost (the ECU will still fuel because the MAF will read higher, but it won't fuel enough to maintain the extra-rich mixtures needed on boost). This could be done fairly cheaply and crudely with a rising rate fuel pressure regulator and high-flow pump. I guess you'd also want to trick the ECU into going open-loop whenever you had boost; a simple way would be to rig a circuit up to throw the TPS reading to wide open when manifold pressure exceeded atmospheric. You also need to find a way to retard timing on boost. I believe MSD has some systems that do this and as a bonus also increase the strength of your spark. Installation of a system like the J&S Safeguard would also be advisable (it's basically a very smart knock sensor and computer). You may want to improve the cooling system, but you'd probably be okay with the stock cooling system. Of course, you'll need good instrumentation -- an oil pressure gauge, a boost gauge, a coolant temperature gauge, an exhaust gas temperature gauge, an air/fuel ratio gauge...

You have to understand that you're pushing the engine beyond its original design limits. You can't stay on boost for extended periods of time because the car's cooling system won't be able to handle it. You have to use the highest octane fuel you can find because the compression ratio is higher than is ideal for boost. You need to be extra-careful about watching gauges for irregularities or abnormalities.

So, you could do it, but it will be work. If you think you could just grab the turbo, intake piping, exhaust piping, and crossmember from a Legacy Turbo or WRX and slap them on your car with a little work on oil and coolant lines, you're probably in for some disappointment, though.

Turboing NA is a by-case basis. In this case, I don't see how it could possibly be better, easier OR cheaper to add boost rather than finding a boosted engine. Rebuilding the block up with turbo parts would still be better but half-a in my opinion because of open deck headers and still the need to keep very limited boost.

If you wanna centrifugal supercharge your Vette engine, heck yeah go for it, chevy doesn't offer a boosted engine. Supra, the block has been proven to be able to handle it (maybe already built for it) and factory turbo engines are more expensive than bentleys. BC/BF Legacies, turbo engines can be had for between $500-2,000 and are MUCH better than the NA block. In fact it is the largest displacement most hardcore engine Subaru has ever made before the new 2.5T semi-closed deck engines came out (and some would still argue about which is stronger). IMO I see absolutely no reason to even consider adding a turbocharger and doing the 800 things you listed to try and make it "safe". Ok, supercharging a EJ22 for a rock-crawler makes sense, otherwise...

Having an NA Subaru that has a chassis begging for boost I have spent alot of time researching this idea. It always comes down to two things. Money and reliability. The fact is: It is cheaper to swap an EJ22T into it instead. Plus the engine will not be able to handle it for long and your need for speed (aka boost control) WILL make you crank it up a little higher until you kill the engine and then it's just the question of why didn't I swap and EJ22T in in the first place.
On the note of 2.5 RS-Ts in the US and Honda Turbo's etc. Those are done for two reasons. One is the lack of an easily swappable turbo engine. The second is laziness. It's harder to tune a japanese turbo engine in an American car than it is to toss on a turbo and call it a day. The engine will not last for long.
I am doing a complete engine and drivetrain swap on my Legacy next winter. I will be racing it in it's current NA FWD setup on the Colorado Hill Climb Association races and maybe a Club Rally or two next year. I plan to put a turbo on it after the end of the season and see how much boost, how hard it can be pushed, on an NA engine and will post my results.

The strongest Subaru engine. Hmmm. Check the MoFo.
http://www.i-speed.us/products/short_blocks.shtml

Two words: Non Interference. Nuff said, I had to replace an engine not knowing whether the valves were bent or not, didn't take a chance and just got a reman. Of course it would have been gobs of money just to find out... I couldn't do it myself then. That piece of mind is worth it to me.

"Two words: Non Interference. Nuff said"

Huh? I'm not a mechanic. I know what non-interference is and I was under the impression that all the early Legacys had non-interference design. Speaking of which I need to change my timing belt. Anyhow, could you explain what that has to do with turboing an EJ22? Like I said, I'm no mechanic yet. Thanks

OK then, I guess I'm gona be guiny pig on this one. Cheap is the name of the game on this one. I can pick up N/A 2.2s for $100 at many places, and I won't feel as bad when I swamp the thing and bend the rods or run it out of oil when I tip the car on it's side, but I want just a little bit more power than the N/A puts out. Either way it's gona have more power than the SPI EA82 that was in it. I'll throw some monitoring equiptment in it and see what I can get out of it before it blows up. Then I'll just toss in another $100 motor and not do whatever I did to blow the first one up.

I think you'll want a turbo ecu. Should be cheap enough on a wrecker. And then run the highest octane you can...

You might want to see if there are some pistons you can drop into the engine to drop compression. Or thicker headgaskets for the same effect.

Realize that neither the original EJ22 ECU nor an EJ22T ECU will properly manage a hybrid EJ22/EJ22T. Due to the different compression ratios, cam profiles, injectors, etc. the engines have different personalities.

I'd almost say you'd be better off piggybacking the original ECU... Or even just doing stuff external to the ECU (like I described above, with a Walbro pump, RRFPR, and J&S SafeGuard to retard timing and watch the knock sensor).

An advantage to not using the EJ22T ECU is that you don't have to add the extra solenoids and sensor, too.

If you're going to turbo an N/A motor, depending on the boost you want to run, you will need some form of computer management to keep your timing in check. Fuel can be taken care of with a RRFPR. Timing curves on n/a motors are more advanced, and with more boost & timing.....it likes to make motors go bye-bye. You take care of the timing advance issue, you won't have any problems, assuming you do everything else properly.

Rod Grosvenor had a turbo'd 2.2 n/a motor, and he was running close to 12-15 psi on it with alcohol injection. He wanted to push it to find out how much it would take. He is in the process of building a 2.2t block....but all his electronics are still n/a. He will be eventually going to unichip piggy back controller, but for the initial break in, I believe he's not changing anything.

Also, talk to Al @ www.torquechip.com His new chip limits timing on aftermarket turbo applications. They learned the hard way. They were logging a car on the dyno, and it let go.....timing was through the roof.

Good luck Subafreak. Let us know how it goes.

Josh - Yeah, you'd need some way to limit timing... The J&S Safeguard is a good safety measure on any aftermarket turbo engine, and it also has built-in boost retard control, so it could take care of that.

To simplify and be safe you'd probably also want to disable the ECU's timing advance, though... unplugging the stock knock sensor would make it use its baseline timing but then you'd get a Check Engine Light. Unplugging the oxygen sensor would also keep a baseline timing, I think, but you'd lose the learning control as well as get a CEL.

I guess Al knows a way to trick the ECU into not learning timing advance while also not throwing a CEL.

Another option would be to switch to a standalone ignition controller. My brother's using a Haltech IG5 that he got for around 200 dollars.

I wish I where in Rhode Island so I could at least watch you do the installation, subafreak, if not help out... Good luck with it.

It sounds like in order to turbo the NA correctly you need a stand alone ecu.

If not stand alone ECU you will do alot of work and end up with little power.

Im going to be the negative one and say not to waste your time. Either that or do it on the cheap to a beater car that you dont care about causing damage too.

It's not so much timing advance you want to do away with, it's too much timing advance.

What the torquechip does is basically limit advance at like 45 degrees or something.....don't have a clue on the correct number. That way you have some advance for bottom end, but as revs & boost build, timing advance will peak out.

Thanks for the ideas guys. From what I'v gathered so far, timming might be my biggest enemy with this but, If it does detect knock won't the ECU back down the timming? I know this is kinda how the J&S safe guard worked. This prodject is going in a beater but it is a beater I need running before snow falls so if I will have to see what happens with the way I got it set up right now and if I blow stuff up I'll just toss another block in there and wire the wast gate open untill I get more cash and time to play with it.

The ECU does retard timing when it detects knock, but knock correction is not as good as having correct timing to begin with, because it retards temporarily and drastically. It wasn't really meant to handle something like turbocharging where intake air temperature can vary widely in short periods of time.

The Safeguard is designed for these kinds of applications though... and (at least certain versions of) it can automatically retard timing on boost.

I believe that if the ECU is in test mode or read memory mode it won't try to advance timing. You may want to leave it in one of those modes at the least.