Corvid

I would double check the advice on here about EA81 rear main seals, since I haven't owned one... But with all the other Subaru engines I've worked on, the rule has been that you never touch the factory rear main seal, because they almost never fail, and the replacement will always leak after install.

For anyone that missed it in the first post, I drilled all the way thru the headbolt. There isn't a coolant passage that I am aware of in jeopardy. What's even better, is that the intake bolt hole is slightly offset to the headbolt through-passage, which means I didn't cross drill the headbolt in the middle, but just barely wiped out one edge of it. What are the odds getting THAT out without snapping it off? If I hadn't ruined the headbolt, this would be a triumphant day. By making a drilling jig to core out the center of the broken intake bolts with a 1/4" drill, I was able to leave just the bolt threads hanging out in the block threads. I twisted a few coils out with a pick and needle nose pliers, and than carefully ran a M8x1.25 BOTTOM tap down the holes. If you do this, use thread cutting fluid and a real High Speed Steel tap, not some grey carbon-steel trash from Ace Hardware. Saved both the holes, pretty as a daisy, no Time Serts.

Oh, I know. This was supposed to be a Weber Swap, and its ending up more like the beach scene in Saving Private Ryan. It's not my first rodeo, but some jobs just have the curse. I'd love someone to explain to me why I haven't compromised the holding strength of the headbolt, and that I should leave it there.

Well kids, let this be a lesson to you. Don't indulge your insomnia by performing critical repairs at 3am, you might forget something small, like resetting your depth marker for drilling out your intake bolts. So, I had the rear intake bolt on each side of an EA82 snap off during careful removal. The stubs are completely chem-welded into the block, and multiple days of point blank heat gun, penetrating oil, left hand drill bits, cold chisel, etc failed. I cut the first one off flat , and noticed that the smaller drill hole was not center enough to keep drilling it out and Time Sert it. I used transfer punches in the manifold and checked their center locations against a factory gasket, and made a drill jig out of a 1 inch block of cold rolled steel on my mill. I bolted it up and it looked perfect. Thats when I failed to remark the depth stop, and plunged through the bottom of the intake bolt. If you are working on the head on a bench, this doesnt matter, because you will drop into air when you hit the passage for the headbolt. On the car however, you can cross drill a 1/4' hole allllllllll the way through the headbolt itself. Then you get to feel very smart. So, my inclination is to pull the engine, pop off the cam carrier, and replace that one damaged headbolt. Anyone with more engine experience see a reason this wont work? Or a reason why I shouldn't have to?

Let this be a lesson to you kids, energy drinks are no substitute for sleep. Surgeons forget tools and sponges inside people all the time, and they are getting paid well, not laying on the concrete in the driveway at night, and doing something else for money all day. Always triple check.

While a gauge may be good protocol, i dont think i need further data on this one. Look what was inside the oil pump.

Hahahaha, its easier to check if i didn't JUST pull the radiator. -sigh- I'll throw it back on and refill, because with my luck, I'd R&R the oil pump again, only to have it be a hose on the snorkus. Thanks.

Well, yes, my concern isnt for the longterm health of the oil pump. My concern is that it seems to not be working correctly and being such a simple object, I'm confused as to what could be wrong. When I fired up the engine after reseal, I primed the oil pump with cranking and the HLAs got quiet, which seems to indicate some oil pressure. Then the OPS started coming on intermitently, so I began looking into things.

Okay, today I resealed the oil pan, pickup tube, and PCV seal at back of oil pan. I also replaced the Oil Pressure Switch. Unplugged coil, cranking never puts out oil pressure light. No identifiable difference, but I'm knocking out options. I got a cheap oil gauge today, but the fitting is wrong, so no data from that. I'd love to hear some ideas about why oil pumps don't do their jobs after a reseal. I'm going to hack into it tomorrow anyway, but really appreciate things to look for or consider. Thanks.

The wiring looks good, no wear spots or contact spots. Tube looks good, no cracks or visual clues of a problem. Not even a spec of RTV in the screen.

I did the typical, cams, crank, oil pump, water pump, seperator plate, access hole, timing kit, valve covers routine. Did not pull the rocker assemblies. Heading out the driveway in a moment to triple check the wire path and the pickup tube.

Yes, I used a factory o-ring. I did not use RTV on the pump, I used Permatex Anaerobic. I haven't check the tube for cracks with a critical eye, I pullled it and wiped it off. Where do the cracks usually form?

I did pull the oil pump, clean it, check the screws (all tight), and reseal it. I haven't gotten a gauge yet, but the loudly clacking HLAs seemed to indicate that its not just the wiring. I also thought it was strange to have the lifters getting oil, change oil and filter, and then hear the lifters starve out and GET loud. I got a new OPS to try when the lifters were quiet, but the parts store kid grabbed me the wrong one, so I just changed the oil and filter first. I don't have a second car, and didn't want to drive back on dry cams so I moved on to pulling the pan and hoped to find an obstruction. I pulled the brand new filter as well, and it was full of brand new clean oil, in case that's usefull to anyone. It seems like the 2 main options left are clogged journals or something wrong with the pump? Something I'm not seeing? Is there anything I could have done wrong R&Ring the pump? I've done a few of these and don't think I did anything out of step, but I always except that I am capable of mistakes. I've been burning the candle pretty hard at both ends, so my mental clarity and creative problem solving are suffering from overwork and undersleep. I'm hoping to crowd source some insight here. Thanks everyone.

Victim is a 1991 legacy, 195K on the ej22, automatic transmission that just got replaced. I resealed the engine while I had it out, except the oil pan since it wasnt leaking. Plugged everything back in, drivers side HLA's got quiet fast, passenger side stayed loud for a while, oil light flickered and acted weird. Next morning, all lifters were quiet, sounded beautiful, still weird on and off oil pressure light, mostly on at low RPMs. I read that a bad oil filter can produce those symptoms, so I got a nice Bosch unit, changed the oil, installed the nice oil filter, and fired up the engine again. Now, with the high quality oil filter on, the oil pressure light was on solidly, and the lifters that had been quiet started ticking. I checked on here, and most people said to check the pickup screen, so I removed the oil pan, and the pickup is perfectly clear. So now I'm feeling tired and stuck. I'd love to hear your ideas.

If you already have the intake off, you've done most of the work already. You will want (2) O-Rings #8069-33010 from the dealership, to install your donor water pipe. While you're in there, I'd look at the black bypass pipe on the drivers side of the engine too, since that other piece of your cooling system rotted off. Of course, for myself, I tend to replace the whole cooling system at the same time if the history isn't known to me. But thats me. And, for the record, there is no listing for that brass nipple in my parts catalog, so I assume its just part of the water pipe.

:ph34r:

Well, a quick check of opposedforces shows 2 oil pans, 31390AA030 for pre-95, and 31390AA060 for 95-on. I noticed a difference in the mounting flange surface when I swapped the pans on mine, but it looked like an engineering decision to eliminate the rubberized gasket, in favor of flange sealant. Presslab said that the 2 different depths of pans had different depths of filter screen pickups and different dipsticks. I already swapped the dipstick over, but I need to know if I have to once again pull and scrape the oil pan I just reinstalled, so that I don't starve the fluid and bork the "new" transmission. I understand that these 4EATs are supposed to be just as durable and reliable or whatever, as a 5MT, but the level of repair annoyance seems WAY higher.

Yeah, I've seen nastier LOOKING rear main seals, which were not leaking, and just covered in clutch filth and oil from the seperator plate. It's from memory, so I may be wrong, but isn't the tolerance for that rear main seal something crazy, like, flatness within .003 of an inch? I am also a huge fan of impact driver on the phillips heads back there. I usually still cam-out the slots on one or two, leaving them stuck in there. The solution is to cut a radial slot in the stripped screwhead with a dremel cutoff wheel, deep enough for the tip of a chisel (or good flathead) laid nearly on its side, and tap the end with a hammer, walking out the dead screw. It's a thing of beauty every time. If you are doing this job correctly, you should only need to end with 2 functional screws for reinstalling the access plate, since the seperator plate should be new and installed with allen bolts (and anaerobic sealant).

I'm creating a new thread because I feel like this question is pretty buried in the one about my entire swap. I started with a dead 1991 4EAT with 4.11 gears. I bought a JDM 1997 4EAT with 4.11 gears, that came on a japanese 2.5l car. The JDM transmission had a dented in oil pan, so i swapped on the one from the 91. Presslab informed me that some of the years of 4EATs had different depths of oil pans, but I am not able to just compare the 2 I have, due to damage and location. I am looking for any and all comparison info, including stamping patterns, cut-off years, dipstick lengths, etc.

This nightmare never seems to end. It was suggested to me that I count the teeth on the ring gear to positively establish the 4.11 final drive ratio. It took a second to figure out, but I stuck a drift punch in one of the holes in an axle stub, fiddled with the gear selector until I found a gear it liked to turn at, and removed the speedometer sensor because it was binding for some reason. The ring gear had a hand-scribed "49" on it, but the teeth counted out at 37, which is exactly what I wanted to see. I reinstalled the speed sensor and checked it, and it seems to work and not bind. I got the 4EAT lowered in and slid back into something like its natural position in the transmission tunnel, resting on a low hydrolic lift table for lack of a transmission jack. That was a headache and when I finally got the bellhousing through the binding spot between the front crossmember and the firewall, I only made it with like half an inch of clearance. So be it. Didn't wreck the wiring harness or breather tubes, and I don't think I ever endangered the wiring for the VSS1. I wanted to get it installed and plumbed so I could put ATF in it and see if anything leaks before I lace up the engine install. It felt good to be close to done. Lost yesterday to heavy rain, and then went out today to install transmission mounts and crossmember, only to have the outermost captive nut on the drivers side snap free. WTF. So I just spent the last hour removing the interior, so that I can drill a %#^@ing hole in the floor and figure out how I want to solve that needless problem, but the rain started up again so now I'm inside on here. Anybody got any opinions on Rivnuts? Anything else you suggest for this situation? If I can get the bolt out of there, leaving the now-free captive nut in the pocket, is there enough room to drill from the bottom and have a Rivnut shove the broken nut out of the way, so that I don't have to drill through the floor?

That is a demoralizing idea. Any advice or cross comparison notes for establishing which oil pan is which? The original 97 JDM pan was dented, so its a little hard to measure. Would comparing the length of the dipsticks work? Do the 2 different pans have a different stamping pattern?

1stsubaruparts.com tends to have the best price and a very wide inventory of factory parts, but expect to wait like 10 for your stuff. rockauto.com is the best source for noncritical stuff, felpro gaskets, etc. most of us grab timing belt kits off ebay from mizumo auto parts (but I don't use the seals, factory only). aftermarket stuff tends to be boutique-built and sold through the forum until you get into mid-90's and newer turbo stuff, where the tuner market takes over.

Hmm. I didn't see any stamped numbers, the 49 looked electro-penned on. It was about the size of a thumbnail, maybe a little bigger, at the base of one of the teeth.

Alright, 37 teeth counted and triple checked. I have no idea why it is scribed "49" on the back of the ring gear itself. Now I just gotta go figure out whats going on with this VSS.

Ok, thanks you guys, got it moving. I tried what crazyeights, gary, and presslab said (thank you for elaborating), but it was still binding. Not wanting to force it, I wondered about the mechanical speedo sender that I plugged in from the 91 transmission... Pulled it out, and now I've got a rotating gear, that reads...49? I thought I was looking for more like 37. I gotta find that chart on here somewhere, and actually count these teeth one at a time.