OB99W

Other than backfire, does it bog/hesitate when accelerated?

Under what conditions; starting, idling, constant load, accelerating, decelerating, etc.? Does gas tank level make a difference? Is the air filter relatively clean (aside from any scorching )?

There are plastic, solvent-resistant putty/taping knives available at most hardware stores or home centers. Also, old credit cards or even the fake ones you get in the mail may work, although some won't hold up to solvents.

Scroll down and check the posts in the "Similar Threads" at the bottom of this screen.

OEM is 38325AA032.

A not-surprising labeling: http://en.wikipedia.org/wiki/Pleiades_(star_cluster)

Yes, there are three places near the top edge where the lower dash panel is clipped on; one each near the left and right corners, and the third just to the left of the steering column's cover. It's usually not difficult to get them to release, but maybe your panel hasn't been out before. You could try tilting the steering wheel to the highest position; that will open a slight gap between the top of the lower dash cover and the bottom of the steering column cover. You might then be able to insert something between the two that will allow you to gently pry the lower dash cover outward. Obviously, to avoid marring the plastic, choose an appropriate tool, and perhaps wrap it with some cloth tape or the like. Once you get the dash cover to release, lower it to the car floor (it will still have things attached). The "3 fuse box" is actually just a small panel with three fuse holders, and is easy to miss. If you get on eye level with the hood release, look above it, in a bit and slightly to the left. The fuse for the seat heaters is the left-most (there may not be fuses in the other two positions). I hope that helps.

I'd never suggest a chemical flush ("qt.flush detergent"). If it's the Wynn's product we're discussing, it's apparently more than a friction modifier, and also not appropriate if the trans doesn't have problems. A slightly older MSDS, giving a bit more detail of the ingredients: http://www.wynns.net/assets/MSD%20Sheets/MSDS%2055701.pdf An short explanation of what the product is designed to deal with: http://www.wynnswa.com.au/menu6_02_09.html Do change the fluid and external filter if it hasn't been done recently, but leave the other chemicals out if this is just for maintenance.

It seems like the power disconnection may have triggered your radio's anti-theft feature. The "CALL" display is probably alerting you to phone your dealer/installer, who can usually tell you how to enter a code that will allow the radio to function as before.

Never have a chemical flush done on a trans, especially one that's not giving trouble. The only kind of flush to possibly consider is one where all the ATF is replaced with fresh fluid. Draining via the plug in the pan requires several refills to approach 100% replacement. However, for a trans that's working well, and if the fluid currently in place doesn't look/smell bad, a single drain of the pan might be enough. What's critical is to get the level right when you're done. Although Subaru doesn't consider it a maintenance item, you might also think about replacing the external spin-on ATF filter. That was something that didn't used to be done much. However, it's become more of an issue in recent years due to the inclusion of ABS on most cars, among other things. Brake fluid absorbs moisture, and eventually can begin causing corrosion. It's not good if that happens anywhere in the brake system, but the ABS hydraulics in particular don't like it. Can't hurt (if done properly ), might not help. It depends what "not so great on the highway" translates to in actual MPG (27 or so?), but 22 "around town" is probably about right, and I question if "much better" is likely. Your driving style, tire inflation, etc., naturally play a roll in gas mileage.

I know a few, actually. However, I believe that any engine oil that meets the car manufacturer's requirements, changed frequently enough (and in conjunction with a good filter), will probably do the job for the vast majority of us.

The 16 blinks indicate an electrically-related problem with the trans, and that further data is stored in the TCU. OBDII isn't the issue; Bserk referred to his "96 Legacy Outback" in the thread you've linked to, and was successful in retrieving the code, so it's likely you should be able to as well. It seems Phase-2 4EATs don't include the code retrieval capability (my own '99 OB doesn't have a wire at pin five of the diagnosis connector), but a '96 should normally be Phase-1. Were you able to locate the under-dash diagnosis connector and ground "probes", and identify pin five of the connector?

Properly classified synthetic oil of the same viscosity rating as "dino" oil is not really "thinner". Previous synthetics did have problems because they were too "pure", and certain seal materials that were designed to swell somewhat in the presence of petroleum wound up shrinking a bit; some current synthetics have more "seal conditioners" (a little "dino" product, etc. ) to help resolve that problem. Also, since synthetics can do a good job of removing sludge and varnish left behind by previous "dino" oil usage, if that removed buildup was at a seal, it might then leak. A not-unbiased source, but interesting: http://www.amsoil.com/articlespr/articlemyths.aspx?zo=0 Many synthetics are mainly PAO (polyalphaolefin) base stock, with additives. Mobil 1 and Amsoil, for example, have a mostly-PAO base. The other primary synthetic oil base is comprised of esters. Certain of the mostly-PAO oils include some esters. Unfortunately, just because an oil container says "synthetic" doesn't mean it's comprised of mainly PAO or esters; the regulations are lax enough to allow some highly-refined petroleum-based product to be classified as "synthetic" or "synthetic-blend".

Yes. In fact, the diagnostic procedure for P0420 says that if certain other codes (P0133 among them) are also present, it isn't necessary to proceed with diagnosing the P0420. As long as you're sure there are no exhaust leaks, P0133 normally indicates a bad front O2 sensor. However, while you're checking for exhaust leaks, look over any electrical connectors down there, and see if any of the ground connections seem corroded. (I'm wondering why the P1442 and the O2 sensor-related codes showed up simultaneously, and an electrical problem might explain that. Of course, it could just be coincidence...)

Another question: Was there any off-roading, or other situation where the bottom of the car could have hit something? (I'm thinking possible damage to the exhaust system and gas tank.)

[quote name='mattoh130']Just before I went for my PA state inspection, my CEL came on. I pulled three codes from it: PO133, P0420 and P1442.[...][/quote]The first two codes could easily be caused by a bad front O2 sensor alone. [quote]I have NO idea where the 1442 fits into this whole thing.[...][/quote]Some info I have for P1442 says "Fuel Level Sensor Circuit Range/Perf", another says "Fuel Tank Pressure System Range /Performance No. 2". That appears to be an EVAP system code. Does your fuel gauge indicate accurately? Did you recently refill the gas tank? Did you possibly overfill it, or leave the cap loose? Any other clues as to recent occurances that might be related?

I'd start by just pulling off the connector at the master cylinder and see what happens. If the problem goes away, it's related to the sensor. If not, look elsewhere (emergency/parking brake switch, chafed wiring, etc.).

In my own experience, that one almost never comes out in one piece on a "rust belt" car, no matter how much penetrant or other "persuaders" are used. In addition to being well stuck, the ones I've seen typically are still solid metal for about one half the original diameter, with the rest rust, making them easy to snap.

It's unlikely that "the girl" decided by herself to warn you about the warranty issue; rather, it's probable that she was told to inform you of that by the service manager, etc. Assuming that some (if not all) the other cars you own(ed) have sticks, it would seem you and your wife would have some experience driving them . I'd suggest having a talk with the service manager, and come to an understanding with that person, rather than the tech; you may even want to write a letter. Otherwise, if the problem recurs, the tech may forget your conversation, or possibly not even be working for the dealership any longer, leaving you in a difficult situation.

It may not be easy to find the problem, even during a "post mortem". It's not uncommon for the generating portion (windings, brushes/slip-rings, rectifier diodes, etc.) to be okay, but for the voltage regulator (built into the alternator) to fail completely or be intermittent.

I know that others (Toyota, for example) do also use the alternator field connection approach. However, it's my understanding that as long as the testing requirement is met, the actual method by which it's done is up to the manufacturer.

Not necessarily; I've seen corrosion on wires that were only a couple of years old, and fine in other respects. The wires can suffer both conductance problems and insulation damage, which is why I said "If the ignition wires are otherwise okay (insulation good, resistance not excessive)". Most of us here generally recommend getting OEM, rather than trying to find a particular brand that's as reliable.

You're welcome. I'm not privy to Subaru's design decisions, but I can surmise based on the wiring diagram. The other dash warning lights already either normally have one end grounded at initial key-on (for example, oil pressure is non-existent, so the sender contact is closed), or are attached to sophisticated circuits that include the test function (example, ABS). The brake warning light normally only comes on with either low fluid or parking brake on; if the fluid isn't low, and the parking brake isn't set, the lamp wouldn't otherwise light at start-up. It seems Subaru was looking for a way to test the lamp that didn't require additional circuitry beyond the diode.

If the ignition wires are otherwise okay (insulation good, resistance not excessive), the ends can be cleaned of corrosion. Use a small wire brush, a knife blade or other metal tool, steel wool (remove any remnants), etc. However, I'd suggest not using "sandpaper" of any kind; the abrasive is usually a semiconductor, and if it remains can cause arcing/pitting of the contact area. Clean the ends of any residue when you're done. Afterwards, before reattaching to the plugs in the engine, take a spare plug and verify that a snug connection can be made at each wire.

NGK may know how to make spark plugs, but either there are typos or the author of the web page the above came from is "confused". A plug that measures 10 ohms from center electrode to shell will be very unlikely to fire; the spark energy will typically be dissipated in the usually much higher resistance of the ignition cable and/or built-in plug resistance, etc. Using an ohmmeter, a new plug will measure infinite resistance from center electrode to shell (ground electrode). Even a plug which measures hundreds of thousands of ohms is somewhat fouled, although it may fire. One measuring tens of thousands of ohms and lower could be significantly impaired. (The ratio between the resistance of the fouling material and the ignition cable or a resistor-plug's internal resistance affects the final result.) Not all dark deposits on a plug are conductive, and therefore wouldn't necessarily be considered fouling. Measure the resistance from center electrode to shell on a suspected plug, and see what it is. However, also be aware that a typical ohmmeter applies very low voltage (typically under 3 volts, although some use slightly more for a high-resistance range) in comparison to the tens of thousands of spark volts; that can mean that a plug which an ohmmeter test indicates not to be fouled may still have a problem under actual use.