hankosolder2

How sure are you about the actual capacity of the tank? I remember Car &Driver complaining about running out of gas in a 1st gen Impreza with a 16 gallon tank, filling the tank after it ran out and it only taking 14 gals. Nathan

Great! That's exactly what I was hoping to hear! Nathan

I was doing some exhaust work on the '98 OBW (5MT, 211K miles) and happened to twist the driveshaft (going to the rear diff) by hand for kicks (just over its range of backlash/free play.) I was pretty surprised to see the front stub axles (correct term? The inner part of where the front CV joints attach to with the roll pins) shifting around (up and down) under the torque. Is it normal to be able to wiggle these around a fair bit, or do I have failing bearings in the trans or something. To clarify- I'm not talking about play in the CVs themselves, and I'm not talking about rotational backlash. The stub axles can be rocked slightly up and down, fore and aft. Trans is quiet with no unusual whining, shifting issues, etc. Nathan

I think that's going to be a MUCH harder swap than going in the other direction- that is, converting an AT car to 5MT. You'd have to install a TCU, tie in to the wiring harness for the main ECM (or whatever subaru's acronym for engine control computer is) add a trans cooler (your rad will probably not have the extra pipes unless it is a universal replacement.) ETC, etc. I'm exclusively a 5MT person , but I think that as autos go, the 4EAT is pretty good- it does downshift on demand, it's pretty smooth and it's a durable tranny given proper maintainence. I'd almost swear that the auto was a bit hotter off the line and quieter at highway speeds- taller gearing when locked up in 4th. It kind of bogs when it hits 2nd gear though. (Admittedly, that's comparing different generations of Legacies.) Nathan

GG, Perhaps the PO kept topping up with water prior to having the HGs done and that accelerated corrosion in the radiator? It sure sounds like a bad radiator to me, and even if it isn't, why not eliminate it as a potential factor? I'm usually not one to throw parts at a problem, but having a fresh rad in a 10 year old car is good for peace of mind anyway. Nathan

Dave, I'm wondering if you're talking an aftermarket exhaust, which sometimes comes in more pieces than OEM? On the '98 OBW, there's definitely only two pieces after the final cat... the long resonator (center pipe?) section and the muffler. I like Honda's exhaust ID system: closest part to the engine is called the "A" pipe, the next piece is the "B" pipe and so on.... Nathan

I was wondering if someone would be willing to confirm my identification of parts in the exhaust system of my wife's '98 OBW. I'm looking for the part which connects the output of the catalytic converter to the muffler... the longest pipe section under there. Is that called the "center pipe" in Subaru speak? Nathan

Re: whistling noise at high rpm/ wide throttle openings. My 98 OBW does that as well. Pretty sure the '95 Legacy also does it! Neither have any misfiring issues. It almost sounds as if it's coming from the MAF sensor in the '98. I don't think it's related to your missing issue. If you had funky plug wires, it could have caused the insulation in the coil pack to break down. Since there are so few parts in the ign. system, like the others said- if you do plugs, wires and coil pack you'll have pretty much "wiped out" the whole system and can turn your diagnosis in another direction if that doesn't solve the issue. I have had injectors open up electrically when hot (i.e. no continuity) but I've never experienced that with a Subaru. Nathan

Well, Nipper, I'm afraid I'm going to have to disagree with you once again! Loss of boost due to booster failure will cause a hard pedal. Less than normal pedal travel with normal pressure applied. Failure due to a bypass leak in the master cylinder will cause a sinking pedal (or pedal just goes to floor.) I'll admit that a non technical person could lump both of those into "almost no brakes, pedal feels funny" which perhaps is what you're getting at? Nathan

Is the brake balance OK? I.e. does it pull to one side or the other on hard braking? I would consider the possibility of something causing intermittent drag and cooking the brakes. (hill holder if it's a 5sp, internally collapsed rubber brake lines acting as a check valve intermittently, etc.) I once had an old BMW which I replaced both front calipers on with rebuilds from Trak auto. I tested the brakes after the repair and they would lock up just fine, nice firm pedal, etc. Drove the car for a week or so, then pulled up to a friend's house and wanted to make a dramatic entrance. Stomped on the brakes as hard as I could and the pedal went to the floor while the car came to a gentle stop. Turns out BOTH front calipers had hairline cracks due to the hamfisted rebuild. The cracks would flex open under forces greater than required to lock the brakes up and the fluid would squirt out! After the stomping induced brake failure, the car "fixed itself" and seemed fine (save for a small puddle of brake fluid under each wheel!) I guess you have to consider freak failures like that. Nathan

Thanks, guys. I already have the diagram thanks to someone who doesn't wish to be thanked! So I will definitely not thank him. I'm going to dig in this weekend on my Dad's Legacy and see if I can't get the shifter bushings and A/C sorted out. Nathan

Does anyone have a PDF or a link to a '95 Legacy wiring diagram? I googled and came up with a bunch o' junk. Specifically, I'm looking at wiring for the cooling fans and HVAC. Thanks in advance, Nathan

First thing is that you NEVER want to flat tow a Subaru with a rope, especially if it's AWD. You can probably get dimensions from Joe Spitz's site. Google Joe Spitz and Subaru... to get the URL. Nathan

+1 the "bump the starter" method sounds dubious, but it works really well. Just make sure your breaker bar is securely wedged. Also, IMPORTANT! There is a revised torque setting for the bolt when you're tightening it. I don't remember what it is, but it's important to torque it more than the spec in the Haynes manual or the pulley may work loose and gack up the keyway for the woodruff key. Do a search here and I bet you can find some suggested torque settings.

That's an interesting theory, Cougar. So you're thinking that the heating element in the O2 sensor is causing a voltage drop somewhere? Makes sense, the heater does draw a bit of current. I suppose you could test that by disconnecting the wires to the heater portion of the 02 sensor and see if the problem went away. It would eventually get up to temp and function normally... Nathan

So, I see that after the last brake job where you forgot to pump up the pedal and totaled your old car you're STILL causing mechanical mayhem! Just razzing you a little...I'm sure you'll get it figured out and working great in no time. Nathan

My attitude on any safety related part is that if a factory replacement is available at a reasonable price... it's best to keep it factory. If some subsequent owner of the car overtorques the helicoil and something goes awry...it could come back to haunt you. It's highly unlikely, but why take a chance? If you've properly installed an OEM part, the liability is on Subaru, not you. Of course, to each their own- it's definitely a matter of opinion. Nathan

That radiator ground wire is right next to the blinker fluid reservoir. kidding aside, what makes you think the radiator should have a ground wire?

You will have to play around with the vacuum lines a little. I don't remember the details, but it's simple. There's no wiring though, and it works great when it's done. Nathan

+1 While any used engine can be a crap shoot, the early EJ 22s were so reliable that barring extreme neglect, you can almost bet that any EJ22 equipped car you find in the junkyard will be there for a reason other than a bad engine. I swapped a used 150K mile one into my car and it has lasted for years and doesn't require any oil top offs between changes! I know you're feeling discouraged, but I think that once you get the engine sorted out you'll be able to truck on for many years with only minor repairs. Remember, there's a 100% change new car payments are going to happen without fail every month. With a used car, you'll take your lumps every now and then, but the average will work out in your favor. Happy motoring! Nathan

Nipper- Are you sure the crank and cam sensors are hall effect on this car? I think they are just straight pick up coils, in which case checking for resistance and shorts to ground is a pretty good test. Most hall effect sensors will have three pins (+5, signal/pulse out, ground) and I'm pretty sure the Subie ones are two pins. Been a long time since I've had to do anything with that, I could be wrong on this. Nathan

I'm going to say that replacing the entire engine with a secondhand one is the way to go here. In addition to the sand, you'll have had all the metal wiped off the cam lobes circulating through the engine. I bet that the oil filter weighs about 5lbs! Of course, if you have more time than money, you might get lucky with heads, an oil pump and flushing the heck out of the engine.... Nathan

If the flexplate is cracked, you have to pull the engine. If the knock is internal to the engine, you have to pull the engine. I guess what I'm saying is that you're going to be pulling the engine no matter what, so get on with it! I suspect you could risk major damage to the flex plate and TC by running it in the way you describe, but I don't know that for a fact. Nathan

Two thoughts- if the EGR passages are restricted somehow (carbon build up?) perhaps defeating the BPT gets enough flow to fool the computer. You could pull the valve and check to see if it's plugged up. Check the diapragm of the EGR valve for a vac leak. Is the exhaust backpressure normal? (i.e. you still have your catalysts?) Nathan

I tend to think that the 1K ohm pull up resistor will be too high. Measure the resistance of an EGR solenoid and use the closest standard value resistor. Depends on what the computer is monitoring. It may measure the current through the solenoid rather than having a simple high-impedence logic pull-up type input. However, I'm pretty sure that won't fix your problem, as I doubt a non-EGR automatic car has a "fakeout" resistor. There'd be no reason for Subaru to include an extra part as opposed to simply telling the ECM not to check that input via either a software difference or grounding/jumpering some input to the ECM. Has anyone tried installing a complete EGR system EXCEPT for the pipe to the exhaust feed in the head? The computer has to be looking for either a change in mixture (via the O2 sensor) to verify EGR flow or a change in the MAP sensor. (Vacuum will drop when the EGR valve is open.) If you allowed the EGR valve to suck a bit of air when activated, I think you might be able to fool the computer. Might have to fiddle with the BPT (as there will be no back pressure to help modulate the EGR flow) and you might have to put an orifice on the input of the EGR valve to regulate the size of the vacuum leak... Nathan