OB99W

Hmmm, that's interesting -- the factory info I have for Subaru P1518 only mentions the condition where solenoid voltage is low during cranking, not high after start. However, even if the solenoid voltage remains high after start, that wouldn't be due to the solenoid sticking. (The wiring diagrams I have show that the ECU is monitoring the solenoid coil voltage, not that to the starter motor itself.)

Meaning you can design multi-link rear suspension to change wheel angles as forces change. Yes, toe is controllable; see "Four-wheel steering", especially third paragraph, in http://en.wikipedia.org/wiki/Steering

I agree with Cougar, it's unlikely that the solenoid or the connection at it is the problem (unless the solenoid is drawing very excessive current, in which case a blown fuse would probably result). The ECU monitors the voltage going to the solenoid, and assuming the battery voltage is sufficient, the ignition switch and the neutral safety switch (on a manual; the inhibiter switch on an auto), or connections to them, are the usual culprits when it's low.

Right/rear quarter doesn't look too bad.

Agreed. I wasn't suggesting that all boxes be opened before they go into stock (except, apparently, certain axles ), but that it would be a good idea for anyone to do that before tearing something apart. For example, I've certainly had "reputable" suppliers occasionally leave a vital part out of a rebuild kit; finding that out when something is disassembled and spread all over the place isn't "fun".

Which is why it's good practice to open boxes and check that parts are correct before starting a job.

Congratulations; I hope the fix is permanent. Good idea on making the changeover to the relay easy for the "little lady" (and yourself ). However... You previously mentioned that there was a click of the starter solenoid heard -- a bad neutral safety switch contact wouldn't typically allow that. It's been warm (for winter!) the last few days, and marginal starter contacts (etc.) often work in that circumstance. Yes, please do post if it fails again, but also if after several weeks things are okay, as confirmation of the repair.

There's lots of good info at The End Wrench site, http://endwrench.com/ . For ABS, in particular look at the archive of the August 2002 issue. These two articles from that issue should allow you to identify which system you have, and help find the self-diagnostic procedure: http://endwrench.com/images/pdfs/IdentifyInfo.pdf http://endwrench.com/images/pdfs/ABS.pdf

Are you really that abrasive? Petroleum will cause many rubber compounds to swell and/or soften. Therefore, lubricants designed for brake calipers don't tend to have a petroleum base, although there may be a limited amount of petroleum content to help keep the rubber parts from hardening. Sil-Glide is silicone-based, and as far as I know, the Sta-Lube product for brakes is mostly synthetic.

You're welcome! I looked at the specs for the Mac ET50 that was suggested above; if you don't need OBD-I or real-time data, that unit does seem to be worth the price.

As you know, the crank and cam position sensors generate pulses based on the reluctors passing by them. It's my understanding that for a certain RPM, the ECU expects each pulse to occur at a particular time. When one is "late" (meaning the engine has slightly slowed), it interprets that as a misfire of the previously-fired cylinder.

Wayneb, it was you who questioned whether valves could have been bent, and I among others told you it was one possibility. See: http://www.ultimatesubaru.org/forum/showthread.php?t=83636 It still is probably a good idea to check compression. Actually, it runs "open-loop" (without certain sensor feedback) when cold, and "closed-loop" when warm. In other words, the ECU uses certain "default" values when the engine is cold; they include things like long injector pulse to enrich the mixture. The rich mixture and other default settings when cold can hide certain problems that become evident when things warm enough for the ECU to go to closed-loop operation.

Gary, before you buy any particular OBD tool, you might consider what is expected of it. The least expensive ones often only read/display stored codes, and allow them to be erased. The best scanners will do that (including the definitions you're asking for), as well as cover OBD-I in addition to OBD-II, display I/M readiness, show pending codes (ones that haven't tripped the CEL yet), grab freeze-frame data (what was happening at the time the code was stored), and allow you to see real-time ("live") sensor data, among other things. They will also cover all the protocols you're likely to see, and be upgradable to future protocols. Medium-priced scan tools will typically include some of those features.

Was the code P0420? If so, the cause is often a marginal front sensor. With the mileage you have, it's worth looking at.

Seems very lucky that the failure became obvious in a parking lot, and not on a highway.

Are all the tires of the same size and properly inflated?

Scroll down and look at the "Similar Threads" near the bottom of the page. One of them reads very much like post #2 above; see: http://www.ultimatesubaru.org/forum/showthread.php?t=24118 I'm not saying that's what you're experiencing, but it bears investigating.

There isn't a common intake cam for cylinders #1 and #2; those cylinders are at the front of the engine, but in opposite banks. The cylinder positions, facing the engine from the front of the car, are:3-4 1-2 If you're certain that the timing is correct now, then it's unfortunately possible that there are damaged valves. As subaru360 suggested, a compression test should reveal that. Although changing the timing belt shouldn't be related to this, a bad ignition cable on either #1 or #2 (due to the wasted spark configuration) could cause misfire on both. However, that would likely show up to a degree at engine speeds above idle as well.

Besides Oregon, I think New Jersey is the only other state that still doesn't allow "self-serve". As far as a mix-up with diesel and gas, it's certainly happened with delivery before; when it does, it affects lots more vehicles than an individual mistake at the pump. http://www.google.com/search?hl=en&q=diesel+gas+mix-up

For most Subarus of the last twenty-or-so years, lug nut torque of about 70 ft-lbs is reasonable; that assumes clean, undamaged threads. For SVXs, somewhat higher (about 85 ft-lbs). Those numbers are "ballpark"; check your owner's manual, which will usually give the acceptable range. Any time you mount a wheel, the torque should be rechecked after about 50-100 miles, since they sometimes loosen a bit.

NOMAD327 covered the topic pretty well. The problem can be caused by tire size mismatch. See: http://www.endwrench.com/pdf/drivetrain/RearAxleBindingInfoF99.pdf If you haven't seen this thread, it includes some pics of what can happen: http://www.ultimatesubaru.org/forum/showthread.php?t=78467

If there has to be complete reliance on perfectly shaped wheel lug holes and the cone taper of lug nuts to center the wheel, then its likely the wheel is improper for the car. There should be a close fit between the hub and the wheel's centerbore, and that shouldn't allow much radial runout. Assuming the hub/centerbore fit is correct (and mounting surfaces are clean), wheel and tire manufacturing tolerances will often contribute more to radial runout than wheel centering will. A typical driver, keeping to "reasonable" road speeds, shouldn't have a problem. Tightening the nuts relatively evenly and close to the correct torque should be sufficient.

Even if anti-sieze was used when the sensor was first installed, the environment it exists in is very harsh; corrosion and carbon buildup are common. If you're lucky, the sensor will come out with reasonable effort. However, I'd suggest being prepared with your favorite rust penetrant. Give the sensor mount a good shot of it, and wait for it to penetrate. If the sensor seems at all reluctant to come out, apply more penetrant and turn it back in a bit before trying to further remove it. You may have to repeat that a few times to minimize the chances of damaging the threads. As has already been mentioned, use anti-sieze on the new sensor's threads and avoid contaminating its tip.

You're quite welcome, Skip. I don't think you're the type to just assume a part is bad, and I didn't at any time say the O2 sensor wasn't bad; it's distinctly possible (likely?) that it is. Although WJM's post triggered my comments, his information is certainly useful. In general, it may well be that a particular OBD trouble code on a particular model almost always has the same cause, and that an experienced tech will on average save the customer money by just replacing that part rather than doing lengthy diagnostics. It's just that I assume many (admittedly, not all) folks using the forum are looking for a diagnostic approach, in order to feel confident that they're not buying an unneeded part (that some may not be able to afford). Another consideration (that WJM's "providing the threads dont self destruct on the way out" brings up) is whether efforts to remove a part will result in a great struggle, cuts/abrasions, and swearing. I hope I'm correct, if that's the case, that people will be willing to spend some time doing the diagnosis. I also sometimes include info in my posts that goes beyond what's needed by the OP, with the thought that others searching later might find it useful. Moving on...

There you go, Skip. It's always good to know which part to throw at a problem. I find it interesting that one dealer tech or another will have what they evidently believe is the definitive answer to a problem, but not speak up until one or more forum members try their hardest to help. If you've got the answer, why wait? Sorry if I was "wasting" anyone's "time" (including my own).