OB99W

Because they want to become members of the "replace the headgaskets" club?

Yes, and there are also mechanical problems that can lead to misfire codes. (It could help to know under what conditions the misfire occurs; engine cold or warm, idling or driven, lightly or heavily loaded, etc.) Nevertheless, spark leakage is common enough to make investigating it worthwhile. We're all doomed to failure in our repair efforts anyway; the "observer effect" (not to be confused with Heisenberg's Uncertainty Principle) implies that looking into a problem changes it.

Simple possibilities are that the radiator fan isn't operating, or the radiator airflow is partially obstructed. When driving, air moves through the grill and past the radiator, making fan operation or a partial obstruction less critical.

There have been several errors in the End Wrench articles concerning measuring coil pack resistance; if that's your point of reference, it's no wonder you're unsure what's correct. The 14 Kohms reading for the secondaries seems about right; my own '99 OB measures slightly below 13 Kohms for each coil. The ~0.7 ohm reading is what you should measure for each primary; that is, primary terminal 1-2 and 2-3 (therefore ~1.4 ohms for 1-3). In addition, you should check from each of the secondaries to both ground and the primaries. The resistance should be extremely high; I'd expect most DVMs to indicate "overload", i.e. infinite resistance. Of course, resistance checks won't detect subtle problems in the coils. You could check for any corrosion at the ends of the secondary wires, the coil terminals, or the plugs, and that the connections are tight. It isn't unknown for either to cause some misfire, especially at idle when the engine is cold. Some misfires are caused by secondary voltage leakage. You can check for that by starting the engine and opening the hood at night; stray sparks should be more obvious that way. Water from a bottle lightly sprayed on the coil and secondary wires can also help in finding underhood high voltage leaks, since a "tight" ignition system usually will tolerate moisture but a leaky one won't.

I wish it were as clear-cut as that. There are situations where the original equipment manufacturer sells the part to a third party (not the car manufacturer), who reboxes it with their own company logo. Since the part isn't being sold by the car manufacturer, and it isn't obvious from the box who the OEM is, the part is considered "aftermarket". However, upon examining the part, it is sometimes obvious that the part is indeed from the OEM. One might wonder why this situation occurs at all; there are two possible factors involved. The first is that the car manufacturer may have contractually required the OEM not to market the part directly. The second is economic -- it's sometimes less expensive for an aftermarket parts company to buy from the OEM source than it is to reverse-engineer and produce an "equivalent".

Obviously, OEM would usually be the "safe" route. However, if you decide to try aftermarket, it may help to know that Unisia JECS (a merger between Atsugi Unisia and Japan Electronic Control System Co.) were OEM for Subaru. Hitachi bought Unisia JECS out. So a knock sensor with those company markings may actually be OEM even if sold in the aftermarket.

There are several possibilities, including a simple one. To help with understanding automotive A/C and responses to your questions, see:http://endwrench.com/images/pdfs/ACTheory.pdf http://endwrench.com/images/pdfs/ACMechanism.pdf Turning the A/C on should cause the magnetic clutch at the front of the compressor to engage. The simple possibility: A very low refrigerant (not "coolant"; that's the stuff in the radiator) level will cause a pressure switch in the receiver/drier to open and not allow the clutch to engage. (A somewhat low refrigerant level may be sufficient to allow the compressor to operate, but provide poor cooling.) Automotive A/C systems are internally lubricated; there is refrigerant oil in the system. However, the A/C should be periodically turned on to distribute that oil. In particular, the shaft seal for the compressor is more prone to leaking if not lubricated, leading to a more-likely loss of refrigerant. Besides the compressor shaft seal, there are tubing fittings that can become leaky due to having loosened from vibration or seals having failed. Repair costs will vary depending on how much refrigerant is needed, leak detection diagnosis, possible seal replacement, etc. That information should be in your owner's manual. The above should get you started. Feel free to ask more questions.

Sorry, I don't have a '90, so can't provide a pic. However, AutoZone's "Component Locator" says: "On your 1990 Subaru Legacy, the Oxygen Sensor is: Under hood, passenger side, lower engine area, below exhaust pipe flange, mounted in exhaust pipe" http://www.autozone.com/servlet/UiBroker/?UseCase=C001&UserAction=performQuickPartType&Parameters=%7C%7EAuto+Store%7C%7ECOMPONENT+LOCATIONS%7C%7EOxygen+Sensor

The ABS system requires a certain minimum voltage level to operate correctly. A failing alternator can lower the voltage enough for the ABS light to be triggered on. Any time the ABS light comes on without an obvious ABS failure, and especially if any other electrical "oddities" are noticed, the battery and charging system should be checked before the ABS itself is assumed defective.

While there are other possible causes, agreed that a sticking ABS motor relay is a common one. See: http://endwrench.com/images/pdfs/ABSInsiderEWFall05.pdf

That's the key to most dent removal; inexperienced people often make the mistake of applying pressure first to the center of the dent. The correct approach is usually to reverse the order of how the dent was formed. Since the center is where the initial impact occurred, it should typically be the last place to be worked. It's also important to have proper lighting, especially if you're working "blindly" from the back of a panel. Even if you can't directly see the end of the tool being used, with good panel lighting and technique you can apply light pressure and determine where the tool is by observing a slight (but not permanent) change in the panels shape. That allows you to relocate the tool properly before applying the force needed to reform the metal.

If a suction cup doesn't work, sometimes a glue-type puller will. Paintless dent removal pros use these (but of better quality, of course): http://www.amazon.com/SIMONIZ-Pops-A-Dent/dp/B00020FTJA/sr=8-1/qid=1171674484/ref=pd_bbs_sr_1/102-5099270-7172104?ie=UTF8&s=hi This type of tool usually works best on small dents. It does require some finesse and following directions.

It may or may not be obvious, but the windshield is an important component relative to overall body integrity -- the car's, and yours if there's ever an accident. Therefore it's very important that you get a good bond; clean the mating surfaces well and follow the directions for using the adhesive (including temperature it can be applied at and the "drive away" cure time before using the car -- cold/dry air substantially lengthens the time needed for moisture-cure urethane). Be careful not to nick the edges of the replacement glass; sometimes what initially looks insignificant can become a crack spreading from the nick towards the center of the windshield over time. You should probably test-fit the glass dry for proper fit and alignment before using the adhesive and mark the alignment with pieces of tape on the glass and body. That way, once you lay down the adhesive bead you can place the glass without having to move it around much.

Heat is produced when electric current flows through a resistance. Corroded/loose connections tend to be resistive. The heat produced is proportional to the power dissipated, which can be calculated by multiplying the resistance times the square of the current going to the headlight. Since headlights draw several amps, it doesn't take much contact resistance in order to develop quite a bit of heat. It doesn't matter whether the poor connection is "hot" (battery plus) or ground, the heating effect only has to do with the magnitude of resistance and current.

Even with a larger-capacity battery and alternator, unless you monitor the battery charge you could still run it down too far to start the engine. You might consider something like this: http://www.jr.com/JRProductPage.process?Product_Id=4027318&JRSource=googlebase.datafeed.WGN+2237 You could either run the computer from it and not use the vehicle's battery at all, or could use the vehicle battery and if it got too low at least have a way of starting. Possible drawbacks I see versus upgrading the battery/alt are either forgetting to charge it beforehand or forgetting to pack it.

I suspect the curiosity being expressed is not so much about wheels and tires but concerning your original statement "I finally got my Legacy wagon back after more than 3 months at AAMCO".

This one goes into my "Weird problems that I'll hopefully never see, but should remember in case I ever do" file. Thanks for the feedback.

Cougar, you've also made some valid points. Note that I began my post with "Assuming the relay itself is at fault", because the problem isn't always due to the relay. Here's a short article that addresses the issue:http://endwrench.com/images/pdfs/ABSInsiderEWFall05.pdf

Assuming the relay itself is at fault, some things I can think of that can cause a relay to "stick" are weakening of the return spring, residual magnetism in the pole piece and/or armature, and roughening of the contacts due to arcing during make/break. If a relay isn't well sealed, oil vapor could find its way to pole piece and armature surfaces; it sometimes doesn't take much of an oil film if one or more of the other causes are also present.

Not necessarily happy, but at least my expectations were met. I've had the problem, to at least some degree; when quite cold, the trans was consistently very reluctant (read took a minute or so) to go into drive, and it was down less than 1/2 pint when measured fully warm. Yes, cold ATF is more viscous than warm, but its volume is also significantly lessened at low temps. In my case at least, that was apparently enough to make the difference. After bringing the fluid up to the proper level, I've never again experienced the problem. Obviously, what approach to take is Mike's decision; I'd be interested in reading of the outcome of the "trans"plant, if Mike would be so kind to post back once it's completed.

I've always found that getting an accurate reading of the 4EAT fluid level is a large pain. First see this: http://www.ultimatesubaru.org/forum/showpost.php?p=479136 Since writing the post I linked to above, I've modified my approach further. Since ATF on the walls of the dipstick tube definitely makes it hard to tell what the true level is, I now pull the dipstick, dry it off, and wait a couple of minutes for what's in the tube to drain back. I then insert the dipstick, wait another minute for it to pick up what is hopefully the correct level, and remove it again to read it. Even after all that, I look at both sides and if one is lower I assume it's the indication that's likely to be more accurate. The 4EAT seems to be particularly sensitive to low ATF level when the temperature is cold; as you can tell from the different markings on the dipstick, ATF volume varies quite a bit with temperature. While you shouldn't overfill the trans, it seems to me that remaining unsure that it's sufficiently filled is pointless if the alternative is to replace the trans. If you truly overfill it, having to drain a bit isn't too terrible. By the way, although too high an ATF level can lead to bubbles in the fluid, too low can also cause aeration. If there's enough air, it raises the apparent level on the dipstick, making you think the level is high enough when it really isn't. Obviously, air doesn't work too well in hydraulics. I'm not saying with any certainty that the problem is due to low fluid, but given you're thinking of changing out the trans, it might be worth taking another look. Besides ATF, other things shrink with cold, so it's possible that some clearance opened up a bit and has reduced hydraulic pressures enough to be the problem. Is there any possibility of putting the car in a warm garage for a day or so and see if that makes any difference? (Yeah, I'm sure that somebody is going to me for offering alternative thoughts to junking the trans. )

Is the tank residue "oily" or "waxy"? Was there an oil sheen on top of the coolant? If not, it makes it less likely that there's a HG problem. One type of deposit can be from corroded aluminum, so make sure you're using a quality antifreeze of the proper type and level of corrosion inhibitors.

Mike, as I'm sure you know, a little knowledge is, umm,... ...so below are links to a lot more knowledge. When you read the following, you might want to pay particular attention to portions concerning High Clutch, Duty Solenoid A, and related line pressure topics. These may take a while to digest, but at least you can do that in the warm comfort of your home and not in the cold driveway. http://endwrench.com/pdf/feb2004pdf/4EAT.pdf http://www.endwrench.com/pdf/drivetrain/02Winter4EATDiagServ.pdf Correction to 4EATDiagServ: http://endwrench.com/images/pdfs/4EATService.pdf http://www.endwrench.com/pdf/drivetrain/02Winter4EATTransOp.pdf EDIT: I reread what I said, and realized that it could be taken the wrong way. In case there's any doubt, I DID NOT mean to impugn the expertise of any forum member; I only meant that Mike might find additional information useful.

I don't know, Skip -- the search function isn't that great, and I've mumbled a few four-letter words, but it doesn't seem to help.

Gas and water don't mix... ...okay, I'll elaborate . I'm not sure exactly what's going on with your car, but here's some food for thought: Fuel tank filling problems are (of course) usually due to obstructions; that can be at the filler, or tank vent. Rust from the filler could have found itself a home in the tank, etc. Water tends to condense in fuel tanks, especially with the temps we've been having in the eastern US lately. If I'm understanding your fueling habits, you're letting the tank get pretty low before refueling; that makes the condensation problem worse. Water in isolated pockets can freeze. A few cans of "dry gas" (anhydrous alcohol) might help if water is a problem, assuming you're not already using gasohol (E10, etc.).