OB99W

Did you have the misfire before changing these parts? After changing them, did the problem go away for a while, then recur? Carbon is often a sign of a too-rich mixture. If there's enough carbon to foul the plug, that alone can cause misfire. What brand and part number plugs are you using? (I'm asking because plugs with too low a temperature range will more readily carbon-foul.) A rich mixture can also be due to a problem with the oxygen sensor. The CTS can cause such problems, and often doesn't generate a trouble code.

Are you having trouble accessing End Wrench?

Yes, Dexron-III/Mercon is correct.

Look's like some suppliers will give you a choice:http://www.maniacelectricmotors.com/13406.html http://www.maniacelectricmotors.com/hiou160ampal6.html

That will probably be sufficient, as long as you don't primarily drive at night with all the accessories on, and/or run an underdrive crank pulley. Which may be why the SVX alt is listed separately.

The '99's connectors are positioned somewhat differently than the pics show, but the harness will probably allow a fit.

I'd describe the '99 alt that way, so it might be a candidate.

Assuming the oil filter is appropriately oriented, it helps to nearly fill a new one before mounting it in order to minimize the time the engine is running "dry" after a change.

The '99 Legacy/Outback appears to use the same 2-pin connector, although the output stud is in a slightly different position. However, the mounting appears to be quite different (as can best be determined from your pics), and I suspect that would be a more important issue in getting alternators from other models to fit. Even the relative position of the pulley might become a belt alignment problem. Unless you're willing to do machine work, it could be worthwhile posting some shots of your alt from other angles, so that the mounting arrangement is obvious.

I'm well aware of the replenishment of additives, etc., by changing out less than the entire amount of the ATF. I didn't say it was necessary to do any specific number of changes, I just provided a way of determining what percentage of the fluid was "fresh" versus the number of changes. Actually, depending on what condition the ATF is in before you start, even a single drain/refill may be sufficient, if done on a reasonable schedule; fluid that has a decent color and smell to it (especially if the trans that isn't exhibiting a particular problem) probably falls into that category. On the other hand, ATF that looks dark and smells burned, or when there's torque bind, etc., probably merits at least 4 drain/refills, if that even makes sense -- at that point doing a full flush is probably more practical, or the trans may even be beyond what fluid change can help.

If you do several ATF drain/refills, here's what you can expect (from a previous post of mine): ------------------- Based on the full capacity of the 4EAT and typical drainage from the pan by just removing the plug (with the car level), I roughly calculated the percentage of old ATF replaced with new by multiple drain/fills. The following is in the form #-of-changes -- %-fresh-ATF: 1 -- 40% 2 -- 64% 3 -- 78% 4 -- 87% 5 -- 92% -------------------

By itself, a sticking caliper won't cause shimmy/shake in the steering. As long as the rotor is of uniform thickness and finish, and not warped, a sticking caliper's symptom will be loss of some gas mileage and perhaps some pulling. If there is sufficient sticking, and it isn't noticed soon enough, that can cause the rotor to overheat, with the possible result of warpage and uneven wear. Once the rotor is in that condition, shimmy/shake are possible. However, at that point the problem will be evident any time the brakes are applied. Since the original poster didn't complain of loss of gas mileage or a consistent shimmy or pulsation with brake application, a sticking caliper wouldn't be a likely cause in this case.

Use of the external spin-on filter (with internal one being a strainer) began on the phase-2 4EAT. That was introduced about June '98, for the '99 model year. Unless your '97 has had its trans replaced with a later one, it would typically be a phase-1, with an internal filter. Have you actually seen an external on your trans?

Confirms what I said in post #6 of this thread: http://www.ultimatesubaru.org/forum/showthread.php?t=89177

The 16 flashes indicate an electrically-related problem was detected, and that there's a code stored. That means the defect can be detected even if the car isn't moved. The B82 diagnostic connector is the same one that has pins for ABS diagnosis. If it's never been used before, it may still be taped to the under-dash harness. It's a 6-pin, usually black, and should have two "ground probes" (separate wires with pins that will fit B82) that are also usually taped.

If the car being discussed in this thread has torque bind, then there's a fair chance that the trans code will relate to the C duty solenoid (blink code "24").

You might try to read the trans code via AT temp light blinking. See bottom of page 12, continuing on page 13, of http://endwrench.com/pdf/feb2004pdf/4EAT.pdf

Besides the need to check the pump sealing (due to both the oil pressure and leakage issues), make sure that the PCV valve and associated hoses aren't gunked up, since excessive crankcase pressure can force oil past seals.

As far as I know, Subaru currently recommends that the Conditioner be used in all phase-2 EJ engines, and recent (well past 2003) owner's and service manuals include that information.

First, congrats on finding the timing problem. Even though the problem is resolved, I thought I'd provide a little info anyway. Obviously, when cylinder #1 is at TDC (Top Dead Center) the mark on the crank pulley will point to 0 degrees on the timing scale. The engine's firing order is 1-3-2-4. Since we're dealing with a four-stroke engine, halfway through the firing order the piston-to-crank relationship repeats, so #2 will also have 0 degree indication at TDC. However, cylinders #3 and #4 will be 180 degrees (half a crank rotation) off from that. The "divide the firing order in half" approach should work on any conventional four-stroke engine. What could change is the number of degrees some of the cylinders will differ from zero at TDC, depending on how many cylinders there are. For example, the Subaru H-6 has a firing order of 1-6-3-2-5-4, so again #1 and #2 will have the 0-degree agreement between TDC and the timing scale. Cylinders #6 and #5 will be 120 degrees later, and #3 and #4 will be 240 degrees. Of course, for each circumstance I described above, whether the valves are closed or not depends on which cycle the cylinder is at. If you're doing a leakdown test, and you have the crank in what should be the right position, but there's no compression, try rotating the crank another 360 degrees.

You're welcome. If the cooling system records are unavailable, or don't indicate Subaru Cooling System Conditioner ( http://www.artsautomotive.com/SubaruCoolingSysCond.jpg ) having been added, you might consider it.

Since it's going to a dealer for a look, why not have them verify that at least the things listed in the following link have been addressed? http://www.autobuyguide.com/2003/12-aut/subaru/impreza/recalls/index.html

The Forester and Impreza are built on the same chassis, and therefore tend to share certain problems.

Radial tire belt separation used to be a problem, and it was suggested that you not reverse the direction of rotation once the tires were in service for any significant time. What Nipper said was the "norm", and I still like to stick with that approach. Most wear differences occur front versus rear (not so much side to side, unless there are alignment problems), so front/back rotation evens out the majority of the problem. However, unless the tires are "directional", it's considered acceptable to cross-rotate current-production tires. See http://www.tirerack.com/tires/tiretech/techpage.jsp?techid=43 and decide for yourself.

For an earlier trans, that's not a bad idea. The external spin-on filter on the phase-2 4EATs does a pretty good job on smaller particles (the internal screen will catch the big ones, but if it does, there's probably serious damage already ). A magnet such as in the Magnefine will sometimes help, but hopefully the 20 miles didn't "liberate" too much ferrous material. Do it while you can.