GeneralDisorder

Lookin good. The sunroof should open about 6" and then stop. Then you have to press the button again to fully open it. I rarely have mine open - it's either too bright for it, too wet for it, or too cold for it. Maybe a couple times each summer does it get fully opened while driving. You'll have fun replacing that roof-line trim. Remove all of it and clean behind it - there will be a lot of crap back under there. GD

We should compare how much we have spent on our '91's . I also got a smokin deal - I traded straight across for a '94 wagon that I had about $1000 into. Came with a low mileage (110k) engine from a '94 TTW half installed and a high volume fuel pump. I figure with the 2007 Forester XT intercooler and the TD05-16G (90* inlet) I have around $1500 into it. Needs a clutch now though after my woman drove it for a month . Totally worth every penny being it's a factory 5 speed car and the interior is pretty sharp still. I have a bit of body work to do eventually on one rear quarter but it's very minor. You'll have to bring it by sometime. GD

The solution is actually really simple - just have someone with a mill oval out the bolt holes for the manifold. Mill down the heads - but only .020" as that is about the limit. It will increase your compression ratio slightly - 2mm off each head would be an ENORMOUSE amount of material (almost .080") and would result is not only pretty serious compresison increase but also possibly requireing custom push-rods to correct the valve train. I say take 1/3 of the distance you need by milling the heads. Take the other 2/3 out of the manifold bolt holes and then port match the intake to the heads. Should work just fine - there is a lot more material on the manifold and head intake ports to work with without disturbing your valve train or compression ratios. Also your exhaust manifold will still bolt up. GD

If you are using flux-core, remember to clean the slag off between weld beads to prevent slag-inclusion in the next pass. If you can get hold of one, this sounds like a job for a stick welder. A larger MIG machine (220v) could handle it, but a small one will make for a lot of work and multiple passes on 1/4" plate. 1/8" would be the limit that I would weld with a 110v MIG machine as the number of passes would become tiresome and quickly eat up the duty cycle of most machines in that size range. If this is a load-bearing component - it might require an annealing step with a torch to prevent cracking from the inherent stresses created in the metal from the welding process. Welding will cause localized hardening and will warp and stress the steel. These can easily lead to cracking down the line if there is flex in that area. Heating the entire welded area and several inches beyond with an oxy-torch and then allowing it to cool slowely will releive the stress and prevent such problems. GD

I must say the EJ22 Brat I just had a hand in building is pretty darn fun. It's setup for a turbo eventually, but it moves pretty nicely already. It's amazing what doubling the HP will do for a light chassis. He hasn't kept track of the total cost of the build, but it was probably around $3000 to $4000 including the cost of the crapped out Brat that was nearly destined for the scrap yard . GD

Welcome to the world of the EA82T owner Angled tip pliers really help out a lot. A pair of 45* and a pair of 90* are very helpful for weird hose locations. Spring clamps rather than screw clamps are best for hoses like that. GD

The difference is that the two sensors are used for ENTIRELY different functions. The front is used for fuel injection mixture control, while the second is used only to check the efficiency of the cat. If you space the front then the mixture control will read a lean signal and dump excess fuel into the engine. If you *don't* space the rear then the cat will look bad. Spacer = lean signal. The lean signal has to occur in the right place or the computer isn't fooled *in the right way*. It's still fooled I suppose - but the joke is on your wallet. GD

Probably not. The computer isn't that smart. It only knows what it see's from the O2 sensors and if that is wrong then it's just going to run like crap. If the sensor's aren't bad then it just has to go with what they are telling it - no matter how rediculous that reading may be. GD

Remove the one from the front sensor - it's going to make it run really rich with that on it. The front sensor is only for mixture adjustment. Someone didn't fully understand what they were doing. It is applied to the rear sensor so that the computer see's a difference between the two despite the cat being bad or missing. This prevents the ECU from throwing a code indicating the cat is bad - which keeps your wallet nice and fat since you don't have to replace the cat at the cost of half your next paycheck. GD

The adjuster bolt is at the bottom of the shoes - just screw the star wheel in a little till the shoes don't rub on the drum. GD

There's nothing to clean and I don't sugest you try to remove them. They can be troublesome and the risk of damage is high. Check their resistance with a multi-meter and visually inspect the tip to make sure it's not worn down - check the gap as OB99W mentions. Mine were about 1000 Ohms with no wear to the tips - the bad one just read open on it's connector. Yes - I used compressed air and a soft wire brush to clean all of mine. GD

Well - the heads differ in their bolt pattern. They simply don't match up. The crank..... there's actually two major generations of the EA71 - narrow case (top starter) and fat case (side starter). The fat case crank might fit. I'm not sure as I haven't really messed with EA71's all that much besides tearing them down for their pistons and such. I have compared the heads at one time or another and EA71 heads definitely will not fit the EA81 and in any case they have tiny valves so it would be a waste. A fat case crank might be compatible with an EA81 crank - I've never seen one but if anything would work I would guess it would be the fat case stuff. GD

There should be 2 ASV's (maybe - but probably.... depends on CA emissions or not). The best way to fix this is to block the pipe's at the heads. What you can do is remove the pipe from the car, and then hacksaw off the end that connects to the spacer under the head - leave about 2" of pipe. Then you hammer the end of the pipe closed - just pinch it off flat. Hammer it hard enough and you won't even have to weld it. Though I usually hammer it and then run a weld bead across the pinch just to insure it's totally closed off. A small leak from that location will not be a big deal though so if you don't have a welder don't sweat it. You might have sucked parts of the melted plastic silencer into the carb as well - so you might have to tear the carb apart and remove all the junk in it. The cat is probably not an issue. If it wasn't clogged before it's likely not clogged now. The ASV failure would not have affected it. GD

Pull the knuckles and buy your bearings/seals and bring them to my garage - I already have a driver tool built specifically for EA series front wheel bearings. Without it, it's a pain to properly install them. GD

A little update: Thanks to all that helped me out with this one - I got a good sensor at the yard thanks to the tip from 2.5_GT, and with the help of OB99W I got the codes pulled and cleared. Thank you both for your time and helpful information. Total cost of repair: $15 for a good used left-rear wheel sensor. The other two codes have not returned - the light is out and the ABS is working. I locked up the wheels in a parking lot and felt it engage. GD

The front hub is pressed into the inner and outer bearing cone's (tapered roller bearings). Pressing it out requires that you press *through* the rollers (they will transmit the force) and as such it requires bearing replacement anytime you remove the hub. It is attached with allen-head cap bolts but unfortunately it only comes off from the inner side of the hub - without removeing the hub from the bearings the hub prevents the tone ring from comming off to the outside, and the knuckle prevents it from comming off to the inside - it is trapped until the hub is pressed free of the bearings. You aren't going to like me after this ...... junk yard. I went and pulled two of them myself. Luckily here on the west coast there is no shortage of cars with ZERO rust. So they come out easily. If there is any amount of rust the sensors are almost always destroyed trying to remove them....... I got lucky on the first try which means I have an extra left rear sensor now which I will sell to you for the $15 I paid for it.... should you need that exact one. I might be convinced to pull one or more of the other's from those same car's for a bit of compensation on the time end of the deal...... Oh - and I just test drove it with the new sensor. Works great. I locked em up in a vacant parking lot and could feel the ABS engage. No light, no codes. So my three codes were really just the code 27. The others were something about a motor/relay and unexpected wheel sensor signals.... probably due to brake dust and gunk build-up which I cleaned away when I inspected each of the sensors. GD

Convert to 6 lug and run toyota rims. GD

Don't ever do the top end of an engine without also doing the bottom. You create a situation where the valves seal like new while the bottom end is worn - this causes stress on already worn bearings and larger amounts of blow-by gasses causing fouling of the engine oil. There is always a high incidence of internal failure shortly after doing a full valve job to an otherwise untouched block. Mill the heads if you must - but only enough to clean the surface. Leave the valves alone unless there is a specific problem that needs to be addressed. Those are general rules. Each situation is different and it's a judgement call. You will have to decide if the condition of the bottom end warrants the higher cylinder pressure's that will result and if that will shorten it's life..... or if that even matters to you. Assuming that's an EJ22T you are working on - that engine *block* is worth $1000. I wouldn't risk it. GD

There's a lot of reasons that people are loyale to the brand... some that come to mind: 1. Really simple to work on - this is one of the big one's for me. The engine layout is not transverse - thus under the hood (especially the older one's) looks like under the hood of an old Chevy truck. All accesories, belts, etc are easily accessed. They are 100% symetrical down the center of the car - thus an axle from the left also fits the right.... etc. It also makes weight distribution very good and improves handling as there is little to no torque steer. 2. Reliability.... generally speaking you can expect a Subaru 4 cylinder boxer to run without any major internal failures for 300k+. I just pulled a part from a car at my local junk yard with 355k on it. It was a '95 or '96 Legacy. There are of course exceptions to every rule and note that I said "major internal failure" - head gaskets have been an issue on some engines for the last 10 years or so. They recently annouced they would be going to the MLM gaskets (turbo style) for all their engines starting in 2010 - that gives hope that head gasket leaking will once again be almost unheard of as it was on the EJ22's of the 90's. Transmisssions are typically very durrable. If the maintenence is done regularly you can expect a LOT of miles from a Subaru. It's commonly said that at 100k they are only just broken in. I have personally torn down 1980's Subaru engines with over 250k miles on them to find the factory cylinder cross-hatching still visible. 3. Economy. The get decent mileage for what they offer (AWD) and the parts are cheap and plentiful. Unlike Volvo or Mercedes, the parts priceing even at the dealer is very reasonable. Even many of us that do our own work buy our parts through our local dealer as the price is very competitive. 4. AWD is what brings a lot of people back. Subaru uses basically the same AWD layout as Audi - they are great in all conditions and have been the official car of the US Ski Team for years. Subaru started their car line with German technology. They bought the rights to the Lloyd designs in the mid 1960's. Ultimately they still use this layout - everything has been redesigned at this point, but the engine/transmission layout is what brings a lot of folks back again and again. It just makes sense from an engineering standpoint - I couldn't possibly own another front-wheel-drive transverse style layout automobile. I couldn't stand it. As for pulling the dash to replace a heater core - that's pretty much the case with all cars and has been for decades. Even American cars are this way. The only thing I can even think of that is not is my '69 GMC pickup - the heater core was almost an afterthought and it's in a box attached to the firewall. But in that truck you could just build a bonfire under the hood next to the battery tray and hire an immigrant to fan it towards the cab . Pretty sure the hood would still close even GD

Yeah - I've never had a need to do it either. The 4 speed's suck and I don't own any 2WD's nor would I bother with them in this area of the country. It's mostly acedemic but apparently there are a few folks that can benefit from the information. Now what I would like to know..... because I'm not sure and have never tried it - will the 5 speed/XT6 pressure plate bolt up to the EA81 225mm flywheel if the flywheel is cut to the XT6 step (0.815")? If so that means that with the 4 speed disc, 5 speed/XT6 pressure plate, and Nissan T/O you could just bolt it all up without having to change the flywheel and transfer timing marks.... GD

All that is correct. But more than starting the car is required for some of the checks. Untill the wheels start turning - some of the diagnostic routines cannot be run by the computer. You will have to drive it to be sure. GD

Yes - you might have to drive a mile or two for the codes to come back - depends on the problem. The ABS unit is continously checking for problems - in my case the left rear sensor was *gone* as far as the computer was concerned - the circuit read open when it should have read about 1000 Ohms. Because of the nature of the failure - the ABS computer refused to turn off the light even after I reset the codes. I reset them (still had the light) then only got code 11. After driving for about 2 to 3 minutes I pulled back into the garage and pulled the codes again - sure enough I got the code 27 (but *only* that code... and 11 of course). I have replaced the sensor and the light immediately went out *on it's own*. I still have to do a test drive and see if there are any remaining problems. I am hopeful it's taken care of. As to the question about cracked tone rings - you actually have to press the hub out of the bearings to get them off - so if that's the problem you are looking at doing a wheel bearing job as well as replacing the tone ring. Newer Subaru's use a tone ring on the axle like a lot of other makes. Though I have never had one crack here on the west coast - I'm sure in the rust belt it's more common. GD

Sweet! My order will be filled sooner than I thought apparently. Ask for a discount. My dealer always gives me a break - should be able to get them for around $75 each. GD

Right - they do all feel different. But the 5 speed/XT6 pressure plate should already be designed to be pushed from the location of the larger 5 speed T/O bearing. The Nissan bearing just facilitates pushing it from the *same* place but using a 4 speed or 2WD transmission and it's associated T/O bearing holder, etc. All the Nissan T/O is doing is helping to put the contact location on the fingers back where it's supposed to be in the first place. GD

It is best, as was your plan, to clear them, test drive it, and then when the light comes on pull the codes and fix what it gives you. Then repeat. Sometimes a single bad component can cause multiple codes - especially if left unrepaired for a long period of time. Phantom codes are not uncommon with stuff like this. GD