OB99W

I looked at the wiring diagram; the motor for the left rear window is wired directly to the switch in its door. Naturally, the switches are set up so that either the main (driver's) switch or the passenger's one can operate the motor in either direction, but they can't "fight" each other. With neither the main nor the passengers switch being operated, the motor has both of its leads grounded. If one of the switches is actuated, one motor lead is grounded and the other is connected to hot; the polarity depends on the direction selected. If the driver and passenger simultaneously select opposite directions, the motor stops, because both motor leads are made hot (and neither is grounded !). The switching/wiring needed to accomplish all of that means a problem with either switch (or wiring between them) can result in neither being able to operate the motor. Sorry nipper, but assuming the daigrams I have are correct, I have to disagree with that part. The diagram for the 2000 doesn't show a lock-out switch. The '99 has one, and it locks out operation of all the windows except the driver's, not only at the individual passenger positions, but at the main (driver's) location as well. It does that by interrupting the ground connection, not the hot lead.

Yes, one connection is to ground, and the ECU won't be happy with the knock sensor dismounted. It will probably give you a P0328 code (and naturally, a CEL to let you know ). If you know what resistance the knock sensor measures, you can try wiring a resistor equal to it between the connector and ground. Apparently some have reported using a 560,000 ohm (560 kohm) one successfully.

As far as I know, SBF1 protects the power windows and door locks circuit; I don't think it would cause a no-start if removed. I suppose you should check the link you moved to SBF2 position and make sure it's still okay. Obviously, if it's now also open, there's a short somewhere on that line. I assume you checked fuse 16 as well.

No matter the "title", if any guy's sense of self is threatened by this, I guess he could always go buy a Hummer . I think I'll keep my OB and ignore any possible implications .

You're welcome, I'm glad that you were able to fix the blower problem with my help in this post : http://www.ultimatesubaru.org/forum/showpost.php?p=505311 You could go to the same link in that post, but this time refer to fig. 20 . Note that if either the 15A fuse (through which the pump relay coil gets power) or the 30A fusible link (through which the relay contacts get power) are open, you'll have neither fuel nor spark. Perhaps during your efforts to repair the blower, one of those got blown/dislodged/etc. Under any circumstance, with the diagram you should be able to figure out what's going on. I hope that helps.

A liter per 4000 kilometers is possibly above average for the 2.5, but not enough so that I'd be particularly concerned. Even when some oil is being burned, you won't see smoke at the tailpipe if the catalytic converter is functioning correctly. Here's what I've said in a previous thread concerning where oil goes and why claims by some that their engine burns no oil isn't quite valid: http://www.ultimatesubaru.org/forum/showpost.php?p=501885 It's often short trips from a cold start that tend to lead to more significant gas dilution of oil, giving the false sense that no oil is being burned. If your trips typically are long enough, and the engine gets fully warmed, there is less dilution and oil that's used (burned) is more evident on the dipstick. As suggested, do check that the PCV valve, etc. are clean and functioning correctly.

While alcohol and silicone are generally safe on most rubber compounds, petroleum-based products usually aren't. There's a reason that petroleum jelly causes softening/swelling; I wouldn't use it unless the alternative was to replace the weatherstripping.

Timintc, if you're referring to the external spin-on filter (like on your 2000?), I believe the '98 doesn't use one. As far as I know, that filter is found on the phase 2 4EAT's, beginning with MY '99.

Lock cylinders get a bit "sloppy" with use; therefore, the sloppiest ones tend to be the driver's door and ignition. Since the cut key works okay on what are probably the "sloppy" locks, but not the "tight" ones, it's likely that the key wasn't cut correctly or cleanly. It would seem that if nothing is obvious when looking at the key, filing is not the answer (how would you know what to file?). I'd suggest going to a dealer who has the correct key blank and have another key cut using key number if you have it (or can get it from a lock cylinder; may have to pull a door panel), or perhaps by VIN info. Make sure the new key is deburred before using it in any locks, as you don't want pieces of metal from the key lodging in the lock cylinders.

Go to this page: http://www.autozone.com/servlet/UiBroker?ForwardPage=az/cds/en_us/0900823d/80/1c/6c/ea/0900823d801c6cea.jsp Click on fig. 44 . The wiring diagram will show you that the switch controls the ground end of the motor circuit (and the grounding of the blower relay coil). The hot end is switched by the blower motor relay. Either the relay is bad, the grn/blk wire isn't getting grounded, or there is a wiring problem to/from the relay. (This assumes that no new problems were introduced by the installation of the new "control setup".)

Unless your weatherstripping is in particularly bad condition, there's something that can be done. Here's what I do about it: Take some isopropyl rubbing alcohol (drug store, 70% kind is good enough) on a cotton rag or decent paper towel and clean the black residue from the glass and weatherstripping. Don't overdo cleaning of the weatherstripping itself, you're just trying to remove loose matter. Get some silicone grease (http://en.wikipedia.org/wiki/Silicone_grease); apply a very thin layer of it to the weatherstripping (so that it won't tend to rub off or transfer much to the glass). The above has worked for me, although you may have to repeat the procedure every so often.

TSB=Technical Service Bulletin There are lots of TSBs, for different makes/models/years/problems. Even assuming we're talking about a Subaru, and a new generation one at that, you'll still need to be a bit more specific. EDIT: Okay, I did a search and think you're looking for info on a sticking clutch problem. The link to an Endwrench article in PDF form was given in this post: http://www.ultimatesubaru.org/forum/showpost.php?p=470501 Click on the link above, and then the one in the post. You'll need Adobe Reader to view the PDF.

You're welcome, I'm glad you got it resolved. Do keep in mind my mention of short life on some lamps; if the rapid flashing begins again in six months or so (I'm speaking from experience), check for a lamp out, and try another brand.

It may well have gotten quieter when you put the fuse in, because that was before you matched the tires. You didn't answer one of my previous questions. To help narrow the problem area, could you try applying the emergency/parking brake in a manner similar to when you used the regular brakes, and see if that has an effect on the noise? New questions: When the noise/vibration occurs, do you feel it in the steering wheel? seat of your pants? Jacking the car up and shaking the wheel might give you a clue, if something is sufficiently loose. However, rotating wheels often is more revealing. There will be things moving other than the wheel and bearings, so comparing sides of the car (left with right, front and then back) may make it easier to tell if something feels or sounds "wrong" (rough, etc.). Some mechanics will lift the car completely off the ground (all four wheels), start the engine, put it in gear, and listen at each wheel. Obviously, that approach has to be done with extreme caution. Unfortunately, lifting the car (with either engine off or running in gear) removes the bearing load and causes the problem to not be very evident with certain types of failure. Please do respond to the questions.

It's key to check the tires' circumference, as others have already suggested, because a significant enough difference between them can be the cause of the problem. See: http://www.endwrench.com/pdf/drivetrain/RearAxleBindingInfoF99.pdf

Well, "in an instant" is just how fast air bags operate on more-modern cars. (No, I'm not directly comparing the efficacy of the tire and bags.) The fiberglass 'Vette body offered little in the way of "crumple zone", so the air-filled donut and wheel rim likely added to crash protection by absorbing energy and/or improving body integrity, and every little bit helps. EDIT: Ahh, NorthWet posted while I composed.

Nipper, I'm curious: Over what period does the scan gauge average the mileage? (It may be real-time, but as you know, nothing is instantaneous, and for gas mileage you probably don't always want it to be.) Have you run a few tanks and compared the calculated mpg with what the scan gauge indicates?

The reason I asked about that is because just like turns cause car body sway and change bearing load, braking also affects play in the same area. It's seeming more likely that there's a wheel bearing problem. To help further isolate the area, does just applying the emergency/parking brake make a difference under the same conditions? By the way, you didn't answer my question concerning whether the FWD fuse in/out still matters.

Evaporator freezing up? As a start, have the refrigerant charge level checked. Also, make sure there are no obstructions to the air flow.

On my '99 OB, the hazard switch button does light, but only when other dash lights are on. Even though the hazard flasher itself works without the ignition key, depressing the switch button when other lights are off doesn't light the button. The glove box light operation seems to be similar, and is one of my pet (minor) peeves. Having to turn on the ignition switch and exterior lights just to get the glove box lit seems a bit strange, although I suppose by wiring it that way a defective or misadjusted box switch can't cause the battery to be drained when the car is parked.

Probably to a minor degree, although I'm not sure that was Subaru's intent. If you anticipate "Somebody" driving your car again, may I suggest a roll cage?

The rate of flash can vary with the amount of current the lamps draw. Depending on the type of flasher, a fast rate could indicate that a turn signal lamp is out; it could be a front or rear one on the side where the rate is incorrect. With the older bimetallic flashers, the rate would increase if the load did. It's possible that an incorrect lamp was installed at some time, which might be drawing more or less current than the correct lamp would. I know you said that "All exterior lights are working", but I'd start by verifying that both front and rear turn signal lamps are operating. Look for a difference in brilliance between the two sides (you could turn on the hazard flasher to make it easier to compare); it could be a clue that either an incorrect lamp was used, or that there may be a partial short. By the way, just because a lamp has the correct type number on it isn't a guarantee that it will work correctly. I've found recently that some replacement lamps are being made very poorly, and don't really meet the specs for current or expected life. It's sometimes worth trying another lamp from a different manufacturer.

Several online dealers sell OEM parts at about 30% off the MSRP (list) price. After discount, knock sensors seem to average about $70 plus S/H; a used one might be "CHEEP", but with that part you could easily get what you pay for.

As 1 Lucky Texan suggested earlier in this thread, torque bind tends to be more noticable in low-speed tight turns. It also doesn't tend to greatly favor turns in one direction over the other. Previously you found that inserting the FWD fuse made a significant change in the noise level -- is that still the case? Does applying the brakes (keeping the road speed in the range that the noise exists) affect the noise?

Uh-oh, I wonder if someone put some incorrect ATF in the trans. See this over on the LegacyGT board: http://legacygt.com/forums/showpost.php?p=256448