GeneralDisorder

I meant purge solenoid, or EGR solenoid - I was really tired when I posted that. Purge and Evap are different terms for the same solenoid. GD

No - it handles evaporated fuel vapor from the tank and lines. GD

Yeah - the valve cover vent ports are probably plugged with carbon. Remove the covers and clean the ports out. Also - did you clean the black plastic "F" shaped connector that connects all the 5/8 hoses with the valve covers, PCV valve, and air box? GD

Definately check the oil pump - frankly if it hasn't been replaced I would just get a new one on general principle. The one from my Hatch looked pretty bad at 224k but the engine is strong and I just replaced the pan + gasket and it had no metal other than the usual amount of ultra-fine shiny flecks. The pump sees pre-filter oil from the pan and the sleeve bearings inside them are aluminium. They usually are REALLY ugly inside after 200k+. If I have my way they get replaced at 100k intervals. As far as oil grade is concerned - why are you running 40 weight? These engines really should be run with 30 weight under most circumstances. The tollerances are very close - they don't get much larger under normal conditions. Usually when the tollerances reach a point where a thicker grade is needed the bearings are about shot and going to a heavier oil isn't going to help. GD

Belt tension is a tricky business. As one of my co-workers like's to say - it's a "1/4 turn before it breaks" sort of thing. As long as it doesn't slip and it runs straight - it's tight enough. Any tighter is just putting uneccesary strain on the bearings - regardless of what a tension gauge tells you. Not a single tech that I work with uses those things - and some of the guys have been doing these for 30 years. Basically you are looking for it to be tight enough to not slip, run true without flopping on the slack side, and make it to the next service interval taking stretch into account. Generally a single adjustment once in the life of the belt is all that's required to accomidate stretch - they only stretch so much and then they will stop. If it makes you feel better by all means use one - but a good eyeball check with it running and an ear for slippage will do just as well. Don't go crazy and pull out your pry-bar and 6 foot cheater to tension your alternator either though - it's a feel and it's something you just have to learn IMO. GD

You don't need the special tools - make sure the slack is on the tensioner side using the crank bolt, give the tensioner a good shove and lock it down. The EA82 belts are crap and don't last much past 50k usually. Tensioned regularly or not - doesn't matter. As such there is no need to be picky about this point. Just make sure the slack it out and call it good. You can easily make a tool - get a peice of 1x1/8 flat bar stock and drill two holes for bolts that will slip into the holes in the cam sprockets. Wrap the other end in electrical tape for a handle. You don't need it though - it's overkill and won't help the belts last any longer. GD

The two on the lines are dampers - the one "on the throttle body" is the fuel pressure regulator. The other item is either the purge solenoid or the evap solenoid. The black plastic cap is a vent dust cap that prevents dust and debris from being sucked into the valve when it changes from open to closed (it vent's the vacuum to atmosphere through the filter cap). GD

Yep - and poor pulley contact will result in over-tightening to prevent slippage, which will cause bearing failure. GD

Now you're just being totally ignorant. First of all - of course they HAVE to say that. These guys are trained to some extent to look for things like EGR valves, and such - they don't know specifics of each model. They go off a basic understanding of emissions devices and principles. If it looks like a duck, and quacks like a duck sort of thing. Yes they will do an "inspection", and they will do a tail-pipe test. As long as it looks good, and performs good they aren't going to be testing solenoids for proper functioning - they wouldn't even know what those solenoids are for, and they won't care. Their knowledge is not that in-depth - they are looking for general proper functioning - they will want to see that there is a CEL lamp that comes on with igntion key-up and goes out when started... ect. Furthermore - if they do pull some kind of Subaru expert out of the woodwork and go through your installation with a fine tooth comb (trust me - that's not how this works), then you are screwed anyway - you have '92 fuel injection on an obviously pushrod engine - there's no way that anyone even remotely versed in EA Subaru's will miss that. And if they do they certainly aren't going to understand that nature of a few solenoids that look to be hooked up as normal. They are looking for big, obvious, red flags. He/she will have a checklist of stuff to look for to make sure it's "in place" (visual verification that it's ON THE CAR). They will check the CEL for functioning. Then the tail-pipe test will tell them all they need to know. As long as it passes, then they simply assume that the stuff they visually verified to be "in place" is working. Don't be scared of these emissions refs - they are people just like us. They don't know everything - chances are they don't know a lot outside of the nepetism that got them their government job in the first place. They aren't going to spend any more time than neccesary hassleing the guy with the mid-80's Subaru. They are expecting to see a transplant of a later model engine into an older car - they expect it to be cobbled together with chewing gum and masking tape - it's what they are used to seeing. Their only concern is their visual checklist, tail-pipe test, and making sure their asses are covered. It's that simple. If you still don't get it.... I can't break it down any more monosylabically than that - some people just can't be taught to think for themselves I guess. GD

O2 sensor's only came on feedback's. So unless the exhaust has been swapped out then you have one. And your mileage sucks, which is a good sign. The computer on the feedback EA82's also controls the fuel pump relay so if it's been removed they would have also had to rewire the ground to the pump relay as well. And no - we can't see the pictures. GD

No - you could write one though I've owned several dozen EA series Subaru's - most of them were Hitachi carbed. Two were feedback - one EA81, and one EA82. For a time I managed to get the EA81 version working fairly well but ripped the whole thing out when I lifted the car and put on a Weber for off-road. That was also my first Subaru back in '98 and I wasn't knowledgable about the system nor had I rebuilt any Hitachi's yet. The EA82 system I wound up with was just an accident - $100 beater that fell in my lap and became my DD. I figured the system would be similar to the EA81 system.... I was pretty wrong of course. The EA82 system is more "evolved" than the primitive EA81 system was. It's got more sensors, more complexity, and the EA82 version of the Hitachi it's hooked to is a demon-child carb. It's actually worse than the EA81 Hitachi in several ways - suffice to say that even though it's bigger and should flow better - I have yet to drive a stock carbed EA82 that was as fast as my stock carbed EA81 despite it having 10 less HP. My advice is this: 1. Buy a used non-feedback EA82 Hitachi from the junk yard - make sure it's got the same transmission type as your's for jetting purposes. While you are there pull every single electric choke housing and find the one with the least spring hook wear. The hooks always wear completely through on the EA82 Hitachi and fall off leaving the choke non-functioning. 2. Buy a rebuild kit. www.thepartsbin.com has the one's I use but a local shop we have here gets them for me for about $20 3. Rebuild your carb using the jets from the junkyard carb, or rebuild the junkyard carb - your choice depending on the condition of each. Don't lose the accelerator pump check ball - it's really tiny, plastic, and will fall out when you tip the carb upside down. 4. Plug the metering ports with vacuum caps - just follow the hoses from the duty solenoids to the appropriate ports on the carb - it's the one's that aren't obviously something else like a fuel supply/return, vent line, or ported vacuum port ect. 10 is about where I put all mine. 12 is a bit high - listen for ping. Also you'll want to readjust your idle speed and mixure if you change your base timing. GD

I don't like them cause the veiwing area is small, and they are expensive. I find a large viewing area and a price of $25 to be more suitible. To each his own though. And if you do a LOT of welding they can still burn you - there is a small fraction of a second before they "go dark" where some UV gets through. Which brings up another good point - don't weld in a t-shirt or shorts for very long or you'll end up with a wicked sun-burn from the UV. GD

No.... no they won't. The smog tests are only allowed to be performed after warm-up. Many emissions devices have to be off durring cold running, and the fuel system will be in open-loop operation. The O2 sensor is ignored and standard fuel/ignition maps are used. The engine will perform WAY outside the testing limits. That's is the nature of internal combustion engines in automotive applications. Emissions tests are always administered when warm - they have to be. Your statement above indicates you still have an incomplete understand of the situation here. GD

The problem is that you are at a different level of mechanical education than "us" (Gloyale, and Myself at least). You think that "as it should" means it must function as some all-knowning Subaru engineer thought it should function two decades ago when it was designed. You haven't yet reached the level where you can look at those engineer's ideas - understand them - and think of NEW ways for those same ideas to still function. When I hear you say (type) something like "as it should" I immediately think "as it should according to whom?". There's a big difference between looking at it your way, and looking at it my way. What you need to understand is that there are a million and one ways to get to the same result, and given 100 engineers you'll get 100 different implementations of the same system. Understanding the GOAL is more important than the details of the implementation. Understand the goal, forget about how it was done by those guys in Japan in the 80's, and get there using your own abilities. Your mind isn't open to the possibility that there is another way. You continue to smack your head against a code 34 wall and wonder why it hurts.... open your mind to another solution. Nail a pillow to the wall - so it wasn't there when the wall was built - does that make it any less useful? Think outside the box man. GD

Two belts is not for insurance - if one breaks the chances are good that it will slip under the good belt and either snap it as well, or pop it off the pulley. It was designed with two belts entirely based on power transfer needs. These are type "A" classical industrial V belts. Individually they have a power transfer of around 2 to 3 HP at around 1000 RPM. A single belt is not sufficient to drive the AC compressor, Power steering pump, and Alternator according to standard v-belt power transfer guidelines. Now in reality, you can run with a single belt all day long and twice on Sunday's because they are over-engineered, and belt technology has progressed since 1985 just a wee bit. But the REASON for the twin belts is purely engineering - there is nothing about redundancy here. If you replace the belts at the reccomended intervals you should never need any redundancy, and besides the only way to truely accomplish that (as I said above) would be to run belt gaurds that prevented broken belts from being sucked into the other pully grooves. I build large compressors at work and am often faced with the questions of how many, how big, how long, and what type of V belt's I will need when building a package - it depends on the size motor and pump I'm running, what RPM I want the pump to turn, the HP and RPM of the motor, ect, ect. GD

And I told you EXACTLY how to do that. You are either not listening, or you don't understand how these two systems work. The EGR code can be eliminated with the resistor, and the EGR can still be enabled by connecting it directly to it's vacuum source. The solenoid only DISABLES the EGR durring warm up - it actually functions as a "de-emissioning" device by turning off the EGR till the engine warms. Removeing the solenoid and plumbing the EGR directly does not bypass any emissions equipment whatsoever. The Purge solenoid functions to pull the evap gasses from the charcoal canistor into the intake. It's a purely passive system, and cannot be tested by the "police". It does not affect tail-pipe emissions except to potentially increase them slightly as evap gasses would enrich the mixture. Installing resistors and leaving the "dead" solenoids in place will be impossible for them to detect in any way, and will eliminate your code, and your CEL, which would automatically fail you..... I don't see a problem here besides your lack of understanding. GD

The carb is far from being the biggest problem with the feedback's. It's the same carb as the non-feedback's but with different jetting. In fact you can take a non-feedback, install the feedback jets and have the same thing - or the other way around. If you have a good feedback, block the metering ports off and install the non-feedback jets and air-bleeds - no more feedback and consistent, good mileage without the expensive, troublesome feedback systems. Why anyone would try to fix the feedback system is really beyond me - other than a poor understanding in general of the difference and how easy it is to switch it over. I've rebuilt plenty of Hitachi's - they aren't terribly difficult once you have done a few. There's no problem with that. The problem is the feedback system itself - which has nothing to do with the carb other than 3 vacuum lines where it connects to the metering ports on it. The metering ports are on all the Hitachi's and are connected variously through vacuum valves and thermo-vacuum valves on the non-feedback installations. If the system applies the incorrect amount of feedback air to the metering ports the mileage or power will be horrible. The best way to deal with it is to just remove the feedback system and jet the carb so it doesn't need it. GD

Hhhhmmm - I hadn't thought of it that way. You are right. I think a lot of us have fallen into that habit as more times than not threads are ressurected accidentally or by the uninitiated trying to answer someone's question from 4 years ago. It then blows up into more people answering the OP's question - wasting their time and the boards bandwidth answering questions for people that will never return for their answers. In this case his question was related, so that's actually a good reason to bring it back. It will still likely result in a bunch of replies answering the OP's question, but such is the way of a forum I suppose. GD

You blew the fuse that supplies power to the fuel pump relay. It's on the same circuit as the radio on some models. GD

Same size I believe - it's the control arm part that's different. This thread is ancient BTW. GD

Here's my write up on it. It's geared toward EA81's - EA82's such as yours are easier, but a lot of the same stuff applies. There's a link at the bottom to a write up by Snowman that was specific to EA82's as well as a partial FSM showing the engine section and specific information on the SPFI system. http://home.comcast.net/~trilinear/EA81_SPFI.html GD

Again - I commend you for you enthusiasm, but if you intend to do something like this yourself, you need to start with a dozen smaller projects and work your way up. Step 1 - get a welder. Get a DECENT welder. A MIG setup with 75/25 gas and .030" wire. Do yourself a big favor and get a 230v machine and install a circuit for it. Get a #9 darkness helment with a large veiwing area - don't bother with the fancy auto-darkening stuff. Step 2 - Practice welding. Start with scraps. Step 3 - Build yourself a welding cart for the machine and the gas bottle. You'll need one anyway and it's a great functional first project. Step 4 - Do many, many more projects. Learn about how metal pulls when it's welded, how to flow your welds properly, and how to run a decent looking bead on different types/thicknesses of material. Steps 1 through 4 will take about 10 years. So stash the Subaru in a safe place for later. GD

Irregarldless, it's probably still not "clogged". You can punch it out and drive on. Most models had only a single cat in the y-pipe anyway. Unfortunately, you have a very difficult situation as I feared from your first post. You have a feedback, computer controlled carb. It's a very complex system - one that is difficult to work with based as much on it's complexity as it's age, and obscurity. It does have some self-diagnostic abilities. There is a computer under your dash on the drivers side. Remove the kick panel and LEAVE IT OFF. The computer is primitive, and does not store codes. When the indicator light on the dash comes on (if it hasn't been disabled) the ECU will flash codes in a standard long-short format. This will help you diagnose the feedback system itself. Most likely, a major cause of your poor mileage is BOTH the feedback system, and the carb itself. You are going to need to rebuild the carb as the top-plate gasket will be shot along with a lot of the rubber bits. It will need cleaning and reassembly to function properly again. You will also need to learn a LOT about the feedback system to repair it and acheive it's rated 28 to 30 MPG. There's almost no one outside of myself and a few others on this board that know anything at all about them - Dealerships are no help, and generic auto shops will do more harm than good in all likelyhood. Frankly your best bet is to yard that carb off and install a non-feedback Hitachi, or re-jet the feedback to the specifications of a non-feedback. Either choice is pretty cheap. You could also install a Weber 32/36 DGV with an adaptor plate, or you could retrofit the later throttle body injection. Both are a bit more work - the TBI (SPFI in Subaru land) being the most work, but also the larges return - best mileage and best drivability. GD

Certainly, but the possibility of that happening can be mitigated almost to a non-concern with proper research into your model selection, and careful inspection of any potential purchase. Gen 1 (90 to 94) Legacy's are without a doubt one of the most reliable vehicles you can find for under $1000. They are just unstoppable. Far, far more reliable than the EA82, and just about as cheap anymore. GD

Double Post!