idosubaru

1. make sure the battery terminals are clean. end clamps are clean and tight. Are they? many times they're not very solid. 2. battery - is it getting fully charged - are these short trips and it's never getting a full charge? fully charge it via a battery charger or another car/jumper cables and test it immediately - what do you get? does the car then start up just fine with a fully charged and tested battery? 3. alternator - test it after verifying connections/terminals/clamps are perfectly clean and tight and see what voltage you get. 4. draw a line across the crank pulley and see if the line separates after some driving. replace crank pulley if it has separated.

Thanks - more and more friends/family/acquaintances/colleagues are asking questions about them. Subaru dealer has one for sale they installed a used transmission into (i didn't know they did that), craptastic 1 month warranty and so far they haven't said what the mileage was.

+1 MT's are worth trying to repair. fewer of them around, warn synchro's, input shaft bearings, clutch maintenance, hydraulic solenoid/lines maintenance, expecting maintenance and repair isn't a bad option on MT's.

Hey, sorry about mom but great to hear you're seeing some improvements.

87 GL 2 door = EA81 EA71 was probably installed a long time ago 3 bolt front strut....a typo or miscommunication...

What’s the overall trend of the CVT failures? Isn’t it more noticeable because of subarus pronounced sales figures or are people avoiding them? What kind of warranty do the Subaru installed units (reman or new) come with?

Sounds like the radiator will do it. You don't appear to have random overheating so that's good. But yes - they can fail without consuming coolant in the initial stages for over a year before finally progressing to more symptoms. The few I've seen just have sporadic, random overheating - usually during times of high demand - high ambient temps, A/C on, etc....even to the point they'd had no overheating events for 6 months (all winter) then start back up again small, minor, random overheating events over the summer with no coolant loss over long periods of time.

Radiator - clean/inspect the fins/make sure nothing is blocking them, and run a hose through the radiator to see if it's restrictive. It should flow full volume of average SFH hose bib. HG. Initial EZ HG's can be variable in presentation - it probably isn't HG but your symptom description by no means precludes them. The heater core bypass is questionable - other explanations are more plausible, or they're outliers. wasn't burped right, hose or clamp that was problematic was repaired during the fix or the issue returned later but the poster never posted back years later when the symptoms persisted or got worse. I've seen H6's take over 2 years to go from initial overheat to confirmed HG failure - that "I think it's fixed" thing is very common on those and EJ25D's. seen it tons of times and it lends itself to poor follow through on internet forums.

Hey Nip! I don’t recommend listening to me with body work but I’ve done it to bumper covers on Toyota’s and Subarus. Take spare bumper scrap or ive even used zip ties as “soldering wire”. Get a soldering iron or other heat device and locally melt the pieces together using the scrap as additional filler.

Reuse. Subaru says it as do well versed Subaru people. Fel Pro and outliers blaze their own trail of replacement take your pick. You only need to know if you replace bolts or not. I wouldn’t think I’m terms such as “TTY”, in a sense it doesn’t matter and unnecessarily muddies up the water.

Leaks that sometimes go unseen are rusty lines above the gas tank in the rear. fluid drips out slowly and just coagulates with all the dirt and dust on top the tank so you never see it. also leaking fluid can build up behind a piston caliper dust seal, but that's not going to take long before it's eventually seen and leaking. Subaru brake lines very rarely fail. I've never seen it happen personally and it's never (rarely) a topic in subaru discussions. Sure - diagnose properly and try to check them but I doubt you'll find your issue here.

common, power issue: 1. check battery 2. clean battery terminals 3. clean battery cable end clamps and make sure they're tight 4. test alternator output 5. makes sure crank pulley isn't separating - draw/paint a line across the face and see if the line is broken after driving 6. check light bulbs - particularly the ones in the rear trunk, etc.

can you tell which side/chain is causing the misalignment? Install one at a time and see if it’s onvious? irrelevancy warning here - but on bike chains, if the chain is compromised it can be necessary to replace the sprockets it engages as well i forget where we ended when I aksed - are chain tensioners under full tension when the engine isn’t running since they require oil supply to function?

Look closely at the outer closest to battery) tab plastic holding the top of the tank, it is actually a clip you can depress. Push that in with a finger and the tank slides right off. Takes two seconds and no tools with your eyes closed.

Subaru MCs rarely fail, hownsure are you that it’s bad? Verify diagnosis first, they’re often erroneously replaced only to find brake issues still exist. since they fail so rarely I doubt anyone but a long term Subaru shop is seeing enough volume to know but I wouldn’t worry much about brand for an MC. Don’t sweat it, it’s so easy on Subarus. They don’t even need bench bled but of course do it, just dont freak out about how it goes or if you’re doing it right. As soon as you turn the thing over to install, fluid goes everywhere leaving air pockets anyway. Unbolt old one, bench bleed, install. About as easy as it gets. If the lines are rusty you may have troubles getting new the fittings disconnected.

That's not a Phase II. But you said it's a Phase II. Which is it? 98 OBS is phase I, use the 98 Forester engine (Phase I) as GD said it's plug and play, you'll want the Forester exhaust mainfold as well as it's dual port and the OBS will be single port. 99 OBS is Phase II get a Phase II EJ25 (Any 00-04 EJ25 or 99 Forester or 99 Impreza RS) and swap it in. Sometimes you have to swap the drivers side cam sprocket and crank sprocket from the old engine onto the new one but that should be zero work since an engine should be getting new timing gear anyway before install. No need for a head swap in either of those, they're both direct bolt in and plug and play.

Two known data points are related - there's no fuel filter and it runs with starting fluid. 04 fuel filter should be in the engine bay - I would be looking a why it doesn't have a filter and go from there. I'm not sure anyone knows what happens to a vehicle without a fuel filter but there's only a pump, lines, dampers, injectors and is there a FPR somewhere on these?

ah right, the pump sock and filter should have gotten most of it, good point.

Original poster take note that says "all Phase 2" but the 1998 Forester you're asking about is a phase 1 so that doesn't answer your question directly, although I think in the end you'll want a Phase II block anyway. Correct - engine mounts are a non-issue for Subaru's, they're basically all located identically, you can drop your engine in a 1987 Loyale or 2008 Tribeca if you wanted to and it'll bolt up. You used an EJ25 HG on an EJ22 block? The fire ring would be larger in diameter (EJ25 bore) than the cylinder (EJ22 bore). Usually the HG matches the bore so the fire ring matches the cylinder diameter. HP estimates are discussed on the board. 190-ish is a commonly thrown around number - but it ends up depending on which block/pistons/heads you use. GD mentions his preferred combination of heads/pistons elsewhere. I think he prefers Phase II pistons/blocks. 1998 forester is a Phase I block. I'm pretty sure Phase II is the prefered block to use but someone else will mention or you can search it. As to manifold and block - the intake manifold doesn't bolt to the block, it bolts to the heads. You're going to bolt your heads to an EJ block - and then your intake manifold bolts to your heads - just like right now. Basically *NOTHING* changes - you're just sliding out the block and pistons and sliding another one underneath all the heads/intake manifold. So there's really zero questions about how to do it - or if the electronics matter - because it's all easy and nothing different happens nor does anything change. Pick a block - bolt your EJ22 heads to it - swap over all your intake manifold, wiring, sensors, etc. Interchanging of intakes/heads/blocks - that is depending on Phase I or Phase II. Phase I intake manifolds only bolt to Phase I heads. Phase II intake manifolds only bolt to Phase II heads. But EJ blocks aren't phase dependent so you can bolt any Phase I or II heads (and therefore intake manifolds because those attach to the heads) to any EJ block, phase of the block does not matter.

Swap in a good used MAF or Idle control valve. They don't fail often but it happens, i've seen both of those fail once or twice. Alternately you could test those components or try to log some live data to properly diagnosis it, but if you cant' do that your mechanic doesn't sound like they will either. It sounds like your mechanic isn't good at diagnosing anyway - they're just guessing, throwing your money at it, and not fixing the problem. Dont' take it back there. There's nothing to "check" - the issue isn't cam/crank sensor related. I think most of us would have told you to test or swap in another MAF or Idle Control valve before those. Get a known good used unit. Members here are very helpful - like the one above who offered a used one. I routinely ship parts to members here and may have a MAF for your car as well if that didn't work out. they're also on eBay.

The chain guide clearance is the clearance between the two guides just as it says - the path the chain travels between the two guides. The alignment pictures I don't recall where I have seen them before while assembling . while I dislike any bit off, if the tensioners aren't under full pressure then a small amount of slack, and a commensurate few degrees off on the pulley alignments, would be expected. I would carefully go over it again, make sure all the pulleys and guides are properly installed, bolts are tight. Go over every mechanical part which locates and rotates the chain and make sure it's secure. If that's all in place then i'd assume you're looking at tensioner slack that goes away once they're apply full pressure. you can probably "mimic" this by visualizing the direction of rotation/slack being taken up and how will impact the sprockets/alignment marks, and on 4 cylinders I have put it under a momentarily increased tension via a sprocket bolt to see what happens, i imagine you may be able to do that with the H6 as well.

Ah - good job catching that. Do the injectors need somehow tested or cleaned as well?

Google “EA81 FSM” or Subaru EA81 FSM. Keep looking I bet they’re out there, just about every other common Subaru for the last 30 years has been readily found online.

That's normal operation of an open differential, they only drive one wheel (so one in the front and one in the rear). FWD = 1 driven wheel 4WD = 2 driven wheels (1 front, 1 rear) With nominal driveline drag you may *incidentally* get rotation of both tires with an open diff. Get an alignment. If you already got one - get one somewhere else that does a better job. There's no shortage of alignment tech's blazing through procedures and not dialing it in. Check tie rods for any looseness If the car had an incident (accident, light hit to a curb, etc) before you got it then the control arm likely needs replaced or possibly the strut. Check them - an alignment tech should pick up on this.

if the issue is fixed it shouldn't need cleared. all the older generations are like that and i've been told newer ones are still the same. the system is looking at real time information and will clear it on it's own if it no longer sees a code. wait a few drive cycles as it may be more than just one data point it's looking at - it may involve an algorithm, database, series, array or some other metric looking at instances within a certain framework that will populate itself out over time.