EJ22t no compression warm

[sciroccoboi]

92 EJ22T in a 90 or so legacy wagon, this is about the worst build I have ever had to deal with and it's been driving me insane.

After having had the wrong bearings in it the first time (friends car, he brought me the crank with rods installed, motor ran for about 20 minutes then the crank ate it for low oil pressure) I rebuild it using a sparecrank and rod set laying around, started it up and it has good oil pressure now.

The car runs rough, misses, etc. I have checked the timing twice counted teeth a billion times to verify timing. Tried different coil packs, swapped injectors plugs, etc. Cylinder 2 seemed to not effect the motor too much and cylinder 4 does nothing when they are unplugged.

Finally did a compression test this morning cold and got 1- 120 3-150 2-150 4-150

I felt pretty good about it so I put it all back together, started it ran it a bit, still no action on cylinder 4 I ran compression again and now 2- 145 and 4 is at like 20. I got upset after that and didnt bother pulling the other plugs.

Could the lifters be over pumping or something? This car is driving me absolutely insane. I should have told my buddy no, but it seemed like an interesting and fairly straightforeward project.

Any suggestions thanks in advance.


Edit: May also be worth mentioning at first I left a bolt out of the rocker assy and that side of the head the lifters weren't pumping up because the bolt was just allowng the oil pessure to shoot out (one of the bolts at either end that holds the end caps inplace on the shaft. Didn't even see it missing, heads were assembled by some previous owner of the motor) so after I put that in it got a lot quieter and ran a bit better ... But I made sure to bleed the lifters and stuff per instructions and leave them soaked in oil before re installing them. This is why I am wondering about lifters.

[Legacy777]

Welcome to the BBS.

Sorry to hear about the issues with the motor. I don't necessarily think the problem is the lifters. If the lifter was collapsed, the valve just wouldn't open as much. Unless the lifter is plugged and can't bleed oil, it really can't over extend. There's usually enough valve spring force to compress the lifter.

You may have a sticking valve, but really you'll need to do a leak down test to see why you're losing compression and where. Short of that, it's kind of a guessing game. You could smell the oil on the dipstick to see if it smell overly like fuel.

[sciroccoboi]

Thanks for the welcome. I figured as such, I was just hoping in the mean time someone would be like oh yeah you dummy you installed a piston upside down or something. Other than that the search continues, I will be checking leakage as soon as I can. I can't seem to find my air valve holder dealy right now and my brothers leakdown tester is in use, being completely broke doesn't allow me to just purchase new equipment. haha.

In the meantime I am going to swap rocker assemblies/lifters from head to head and see if it makes a difference. Or if they cant be swapped just swap the lifters. Figure maybe somehow they are not bleeding off.

[sciroccoboi]

Swapped rocker assemblies, no change. I'll put air into the cylinder pretty soon and see what I can come up with.

[sciroccoboi]

So I got ahold of leakdown tester and it doesn't show any excessive leakage, anything that is leaking is from the crankcase. (but only like 20-30 percent when the pistons is somewhere in its compression/power stroke)

The car wont run now though. I will start and fire, and pop and sputter and run at like 500 rpm really harsh and hiccup and then die, if I move the pedal around while its doing it it seems to stay running as long as I don't hold the pedal in one spot.

Does the cam sensor fire the injectors in these cars? I'm wondering if its faulty and washing out the cylinders? Injector o rings looked okay when I swapped injectors. but I noticed when I ran it before just now that it was glowing the cat. Can a cam sensor cause these issues?

Also just to make sure I have an na cam gear on it off a random legacy in the junkyard, looked to be of the same era. are they the same as a 22t?

Oil doesn't smell like fuel or anything. It's slightly overfull but I'm doubting that is causing all these problems. I'm lost. THis is my first engine build that hasn't worked right in 5 complete engine builds.

[Alphius]

Any codes stored?

To me it sounds like a sensor/electrical issue more than a mechanically dead cylinder.

Cam gear from an N/A should be fine.

[sciroccoboi]

Yeah, the last I checked the only code was for IAC.

I'm thinking it has to be some sort of sensor as well. Fwiw the reason it wouldnt start right after my leakdown test was because I had the intake tube off and forgot to put it back on, haha.

I held the throttle at around 3k and unplugged the cam sensor and the motor didn't run ANY different. Is this normal because i would think it should take a fat dump or at least run somewhat differently. I'm going to try a different cam sensor tomorrow from the junkyard.

Right now it has the NA ecu in with known good injectors. I drove it for a while with the NA ecu and pinks then finially swapped all the pins to run the turbo ecu. I dont remember but it could have started this at that point but it never has run properly, I thought it was because the NA ecu hated the pinks but swapping out for the na injectors didnt do anything with either ecu.

It pops and backfires occasionally on steady throttle and whatnot. Sometimes seems liek the RPM wavers a little bit, i.e. smooths out and then gets rough, raises and lowers without any throttle input. Either way 4 is dead to the world no matter what.

[Alphius]

Do you have spark and fuel on 4?

[sciroccoboi]

There's spark, and the injector gets a pulse near as I can tell (smart probe on the ground wire gets a pulse just like the other cylinders) so unless I have got crap injectors in there twice. after switching them around, and replacing them with a good set, there shoudl be fuel.

And it does have power ... and putting a screwdriver to it you can hear it functioning.

[Legacy777]

You're losing compression somewhere....and if the leak rate on that cylinder is no greater than the others than I'd be suspcious about the test. Did you leave the coolant cap off when you were doing the leak down to see if any bubbles came out through the cooling system?

Does the timing gear you grabbed look like the one on the left in this picture?

http://www.main.experiencetherave.com/s ... P_4381.JPG

How long was the engine running to get the cat glowing? Is this a stock cat? There must be quite a bit of fuel in that exhaust/cat for it to glow...

You really need to run the ej22t ECU and get everything setup properly to do your troubleshooting. You can't run the pink ej22t injectors with the non-turbo ECU and expect it to run properly. Also, as it sounds like you're aware, but in case you're not, the cam & crank sensor wiring is swapped between the non-turbo & turbo ECU's. You should be able to just swap the positive pins at the ECU connector.

[sciroccoboi]

Yeah, thats the cam gear, I had two no reluctor gears when putting it together but I noticed right away and fixed that before installing the belt (guy brought me the motor with two block haves, two heads, and a couple boxes of parts, duplicates of everything, some triplicates. All different in some way. But two passenger cam gears, haha)

Yeah, I had the cap off and nothing in the cooling system. I'm starting to wonder if it was a freak accident. And right now it's running the NA ecu with NA injectors. So the only thing different from when the car itself ran with factory motor is the IAC valve.

Yes, its a stock cat, and probably about 4 minutes total with half of that spent somewhere between 2-5k varying all over. That was my thought, it has to be a LOT of fuel to get the cat to glow. Unless the car has to be really hot, I'm going to run the leakdown again this morning. It just doesn't make sense. Yesterday morning I was REALLY hoping for a broken cam or severely pumped up lifters.

I switched back to the NA ecu becuase I don't have a turbo map sensor or wastegate control solenoid at this point. Figured maybe the ECU didn't like it and was having problems. I only tried the turbo ECU for less than a couple hours, then I figured out that it has much deeper issues. I don't want to tear the motor apart a third time. I'm hoping it's just a sensor issue and the guage lied to me about the compression.

[sciroccoboi]

Did a leakdown when engine was fully hot and am getting 50 percent leakage into case. Guess I'll be removing the motor again and checking the piston rings, maybe one is broken, or something. I don't know what else to thikn at this point. Replaced cam sensor, knock sensor, and plug wires (all were stupid cheap at the junkyard so I figured getting a nice clean set of osmething couldn't hurt.

So I'm going to go start pullng it ... again. ANd then pulling the head off ... again.

[Alphius]

Should still run and idle pretty good with 50% leakdown in only one cylinder... Hope you get it running after a teardown! My bet is on rings or a ringland.

I had a cylinder wall gouge in an EJ22T once that ran pretty good, the only clue was a flutter in the vac gauge at idle and a little rhythmic roughness at idle. Off idle it smoothed right up. Had about 40% leakdown on that one, cylinder 3.

[sciroccoboi]

Okay so here's what I just did. Sorry to post a million times but I want people to know what happened becuase if anyone is running into this issue I know it would sure have helped me a lot to see something like this haha.

So I took the rocker assembly out of the drivers head again, smashed the lifters in a pair of channel locks until they were flattened out. Then I did NOT bleed them (left them nice and squishy)

reassembled and left the same bolt out of the rocker assemblies that I did the first time I built the motor so they couldn't pump up again.

The motor ran TONS better. And unplugging cylinder 4 injector now makes a noticable difference as well as cylinder 2 which previously didn't make too much difference.

When I was installing the rockers again I pushed down on all of them to make sure there was lash and the rear most intake valve on cylinder 4 didnt have anywhere near as much lash as the others.

My thought:

Whatever machine shop did the valve grind didn't check the valve heights after grind and just assumed they were okay. Once warm the valve is too long and cant seat anymore if there is oil in the lifter.

It wouldn't surprise me as the valves still had grinding compound on them when I went to install the heads on the motor, had to clean it all out. If the machinist didn't even remove his lapping compound why would he have checked install height?

Sound reasonable?

[ciper]

How weird. They should adjust themselves. My wild guess is metal shavings have made the lifters jam up.

[sciroccoboi]

Im going to pull the orings off and soak them in carb clean overnight and blow them out. Will see. I'm hoping its just lifters, I'm tired of this car, haha, and western washington hasn't had the most friendly weather for working on cars outside, even with my ezup.

Edit: If this fails I may install the roller cam/manual adjust rocker setup out of this NA ej that's sitting next to my house just as a last ditch effort, replacing the cams is still better than pullng the motor apart

[sciroccoboi]

Ring gap when I pulled the piston out. Is this a possible cause? No cracks, no nothing anywhere. Took rings off and measured ring gap and they are HUGE when installed in bore. way over spec but below the service limit at least. I suspect the others are the same.

The owner of the engine also had the rings installed on the pistons when I broght them over, I inquired as to if he had mesured the end gap on the rings and he told me they (him and his machinist) had. So I didn't take the time to re check, didn't want to risk breaking a ring somehow.

However his machinist also told him that the block was perfectly fine, didn't even need to be honed, just throw new rings in and bolt it all together ... The bores are obviously wiped out, have ledges in (likely) all of them. And to recommend just putting new rings in without honing? Is subaru special, because all I have ever known is that is a No no, at least not if you expect longevity with good compression. I honed the block lightly before I installed the pistons because I wasn't going to let that one slide by.





Bores are pretty hammered, there's a small bit of corrosion that I don't know how it got there but it's not bad really, I've seen motors with much worse and great compression. The'res a lip a couple inches down from the top on the bottom half of the cylinder but its there for bore #2 as well and I think all the bores. Buddy told me he didn't have the money to bore it out and get pistons, or get it re sleeved or whatever, was told it would be fine. I said it should run okay anyways, and built the block. The other cylinders have compression, the only thing I can see with this is the ring alignment. I'm going to put the gaps back differently and put it back together.

I figure if it's ring gap it could explain why after each deflation of the lifters it runs well, until I drive it or it warms up real fully. Cylinder expans, ring gap opens a bit more, and compression just says bye bye? I'm guessing here. Headgasket was fine, no visible cracks in liner, no cracked ringlands, just the rings, and the bore. Heck the bores don't even have scoring really.

I replced the rockers with another set out of his spare turbo heads he had and it was running like a champ, lifters were loud because they were all deflated and un bled, haha. It warmed up, did all good, then I drove it. After the second or so time I hit the gas it started running poorly again, wouldn't go past 4000 rpm easily, sputtering etc. Pulled compression and had 30 psi on cylinder 4 and high numbers in the rest. Checked and the intake lifters hadn't even pumped up for that cylinder so I knew it wasn't valves. Leakdown showed nothing out the exhaust, or intake valves, Just the crank case.

5 minutes later compression was back up to 85 psi in that cylinder. I pulled the rockers and still got high leakage into case. At which point I pulled the motor out. (36 minutes start to hanging in the air with no air tools, personal best, haha.)

[Alphius]

Wow, sounds like a shoddy machinist.

If it was me, I would lightly hone those, fix the ring gaps and check bore size vs. piston size. Then check valve sealing and possibly lightly lap the vavles to the heads again.

I bet you just have a combination of problems, and fixing everything up will make it run much better.

[sciroccoboi]

How can I fix the ring gap if the rings are already too short? I've always been able to file my rings down to spec, never had the issue of them being too smal to begin with, haha.

The issue is really that my guy cant afford to spend money. He's over budget on it right now (had a respectable one too I think) and he's trying to buy a house, getting married, etc.

I was going to touch up the bore with my hone again but I don't want to cause an already bad problem to get worse as far as ring gap.

I'm trying to get ahold of a bore guage, anyone know a trick to make one?

I'll fill the ports with water and see if anything drips out. even if it doesnt I might lap the valves lightly just to be sure. How great a job could have been done really when the machinist left lapping compound on the valves after he was finished?

Also I don't know if the person who built the heads is the same machinist as the one who told him the block was good to go. Either way it's all just fishy.

[Alphius]

That ring gap is a stumper, you're right. The question is are they too short because someone filed them too far or are they too short because they're some poorly made Chinese rings or something? The only way you could fix that problem fully would be to buy new rings I think. If you get them clocked right, you might have better luck anyway. The way your picture shows could definitely cause some blowby issues.

Check this bore gauge out, I've done similar in the past, and while not perfectly accurate I think you can get close enough to make a judgement call in your case if you can't get ahold of a real bore gauge. I hope you have a set of calipers. http://www.ajs-matchless.com/TechTips%5 ... egauge.pdf

Water usually has too much surface tension in my experience to check valves, use some rubbing alcohol dyed red or kerosene since they will show even the smallest sealing problems and won't hold tension like water.

It does sound fishy. I'm with you there. On a budget, I'd say the best thing is to do like you're planning, measure up the bores and pistons, clock the rings better, give it a light hone and hope for the best. It would be nice if you could check valve height and fix that like you were talking about before as well. However, if compression is going into the crankcase, that shouldn't be where your problem is. I don't recall if you ever had valve leakage in your testing so far?

Check this too, I wonder if you have slightly larger bores than you have pistons and you're over tolerance on them? Maybe at least on number 4 where you're losing compression. This is why I think it might be worth measuring pistons and bores and see if you can match up smallest bore to smallest piston etc. if you come across any irregularities.
http://bbs.legacycentral.org/viewtopic.php?f=5&t=40487

Keep us posted, I'm interested to see how this turns out.

[sciroccoboi]

THat bore guage thing is pretty legit actually haha. ANd the piston is a b size piston, all the bores are b. However it seems that the piston is under spec, unles my calipers are wrong. But it could be. However if I can get ahold of a bore guage I might find that the other piston is a better fit in this bore or something. Maybe if the other bore isn't as worn I can swap the pistons side to side and come up with some respectable ring gaps. I had wondered about the piston sizes myself actually when I saw in the repair manual about the stamping on the block with the
size abc etc.

Well, I did have the rings that came off the pistons in the first place, I assume them to be O.E. or at least dealer rings. They are a bit thicker and I managed to find a different pair of compression rings out of them for the bore. The thing is that they look literally brand new. I'm tempted to just run them as they are thicker and of significantly higher quality.

The rings he used are DNJ PR708 Standard. I found out they are a relatively cheap brand it seems. ANd they actually look more worn than the rings that he replaced, even with the minimal amount of run time. Though to be fair the motor did suffer major bearing failure on the first build, and the amount of metal in the oil could have contributed.

Unfortunately I one of the second rings is broken from the oe set, but my temptation is to use as many of the old rings as possible, then use the new set's oil control rings.

I'll check the valve height too, but I never had leakage through valves at all, not once that was detectable. In my experience its pretty hard to actually have a valve not seat even with the most worthless lapping/grind job ever. If I can make my 270,000 mile b2200 valves seat and seal just by lapping, I'd imaigine these cant be much harder. haha.

The biggest annoyance is basically that I have to re do all the checking that had supposedly been done before. This has been a nightmare but I'm feeling better about it because the more I check into things the more I'm feeling like I didn't have a chance of success in the first place. haha.

As soon as autozone gets the headgasket in I will slap it back together and hope for the best, if it doesn't work out I already told him it will need rebore and new pistons/rings to work out, period. At this point I would just be happy with it running, forget the turbo, haha.

Thanks a lot for all the help guys, I hate to be the guy who just makes an account to ask for help, but sometimes ya have no other choice. Hoping I don't seem too idiotic or hopeless here.

-Royce

[Legacy777]

Just out of curiousity, have you pulled the piston rings and checked them in the piston bore? If so, what did they measure out to be?

[sciroccoboi]

I did, that's when i found the massive ring gaps. I can't give you the exact numbers at this time as the car is far far away, but they were way big.

The original rings that were on the pistons offered tighter gaps. Or I found a few that did. I'm still going to have to measure bore and stuff.

Works or not at this point it's probably going to wind up needing a bore job. Or maybe a piston. I don't have all the proper tools to measure it yet.

[sciroccoboi]

Well boys, I threw it back together with a better fitting set of rings for #4, a light hone, and a new gasket.

Started it up and it runs okay, till it gets warm, then there is no compression on 2 or 4. I don't get it. I will check cam timing but i think the motor is just done for. I told my friend I am giving up and he can have someone else do it as clearly I am incompetent, haha. I guess I should stick with inline/v engines.

So, with a depressed attitude and heavy heart, I suppose we can put this one to bed, least for me.

Thanks for the help guys. Im suspecting the pistons/bores are just too warn, plain and simple.

[Legacy777]

Well that sucks....but I totally understand you wanting to move on.