OB99W

It is ''related'' (although wasn't the cause in this case), and that's why I asked Megell (Mike) whether the dash warning lamps were operating correctly. In addition, it's possible for a blown fuse to prevent a good alt from charging a good battery -- the dash lights and certain electrically-powered accessories also wouldn't work then, but might be assumed to be symptoms of an alt failure, rather than that blown fuse being checked for.

There are two nuts that have to be removed. In order to access them, you'll need to remove both the upper and lower service covers. Once the nuts are removed, the lamp assembly has to be pulled straight back to release it. I could provide the rest of the details, but the owner's manual has the info in the ''Maintenance and service'' section.

You're welcome, although I hope you won't have to apply the info anytime soon.

Hmmm, my brain apparently fell asleep when I initially answered your question. Besides the crank sensor, the other item easily seen when the alt is out is the oil pressure switch. The two coolant temp sensors are on the coolant pipe, and kind of buried. The one with a single-pin connector is the one for the gauge (referenced to ground), and the 2-pin one is connected to the ECU.

I'm not sure exactly what you were looking at, but the two might have been the coolant temperature sensor and the crankshaft position sensor.

The problem with warning lights is that they tend to be ''go/no-go'' indicators. In the case of the charge light, the alternator output has to drop sufficiently below the battery voltage before it will be lit. That means in some cases even if the output isn't enough to charge the battery, it still might keep the charge warning light from coming on. A dash-mounted ammeter would be much more revealing, but a voltmeter could tell you enough in most situations. Unfortunately, the ''average'' driver wouldn't know what reading was acceptable and what wasn't under various conditions.

Since it appears to be engine oil, I'd suggest checking the PCV system (valve and tubing) for proper operation. A malfunctioning PCV can allow engine blowby to pressurize the crankcase, and oil might then exit from an imperfect seal.

That's not the typical alternator failure that we see, although the alt could still be bad. Could you start the car once more, and measure the voltage (to ground) at the alternator's output terminal (not at the battery)?

It's obvious that the alt has insufficient or no output. Alternator field current initially comes from the battery (until the alt puts out enough), through the dash charge warning light. If the charge light and the brake warning light are lit with the engine running, that's a pretty sure sign that the alternator is bad. However, if there's no alternator output, but the charge light is not lit (as is apparently the case) with the engine running, then the alternator itself isn't necessarily the cause. Before you condemn the alternator, please tell us whether the charge light and the brake light come on when you turn the ignition switch to the ''ON'' position (but don't start the engine). Based on your answer, I might have other questions or suggestions of things to check.

I'm sure that there are people who'll tell you they gained 5 WHP or so by replumbing the intake, but it doesn't come without consequences. As Nipper already mentioned, the cars are pretty well engineered to get what performance that can be had out of the engine and still have it be reliable, etc. I'm not going to suggest that Subaru (or any other manufacturer's) engineers build perfectly efficient cars. One of the reasons that some companies participate in racing programs and/or research at universities is that they sometimes learn a new trick or two. However, engineers tend to be aware of what aftermarket companies are doing, so if it were cheap and easy to get the 15-20 HP you're seeking, it would have likely been incorporated into the design.

I only mentioned it because I thought someone might think there was a trick to sealing a leaky exhaust connection using those parts.

It seems that Guinness133 is talking about an intake (not exhaust) ''resonator'' (silencer). That part not only reduces noise, but it also minimizes intake air turbulence that can occur at certain engine RPMs. Eliminating it can therefore cause air metering to be inaccurate under some conditions, causing other problems.

Looks like you had a camshaft to seal and a starter that needed contacts as well.

The flywheel and clutch cover (pressure plate) balance marks aren't supposed to be offset by precisely 120 degrees, but rather by more than 120 degrees. Ideally they would be offset by 180 degrees, assuming the bolt pattern would allow it.

It's fine the way it is. Those double marks are not that critical. See the next-to-the-last paragraph in http://www.motor.com/magazine/pdfs/102001_08.pdf (and ignore the flipped-over photo ). While it's obviously best if the alignment is perfect, even if there was a misalignment of the more-critical single marks by 1/4 tooth, that would only be about 3-4 degrees off, or 1%.

211, I hope you don't mind, but I'm going to change the order of what you wrote so that my comments are more relevant. I've found several errors in Subaru literature (both factory and other) concerning resistance measurements of coil packs. Specifically, the primary can't be tested with an ohmmeter on packs having an integrated igniter. Insufficient spark can cause a misfire that doesn't show up at idle, but will at even slightly higher RPM. That can be evident as a rhythmic drop in the vacuum reading each time the ''bad'' cylinder misfires. There are things other than ignition problems that can cause a similar vacuum drop, but the weak spark from the #4 wire you mentioned is something to further investigate. The coil would have been my concern as well, if you hadn't already swapped it out for one that assumedly is known to be good. If I read things correctly, you also already swapped ignition wires. I'm beginning to wonder if the ECU is getting a particularly weak pulse for cylinder #4 due to either a bad position sensor or damaged reluctor.

For a particular temperature and pressure, the number of molecules in a volume of gas is a constant, and therefore humid air is less dense than dry air. (Hydrogen from the water has less mass than other air molecules.) Since the presence of water vapor displaces some portion of the other components of air, there isn't as much oxygen available to the engine for combustion. The mixture is then richer than it would be with dry air, and for many vehicles that results in ''better'' performance.

There have been some reports here of the bolt shearing off on 2.2s at about 125 ft-lb, which is why I said 140 ft-lb could be ''risky''. It's possible the bolts that broke were already stressed. More to the point, even 115 ft-lb should easily be sufficient to seat the crank pulley if nothing else is wrong, so if there's still a wobble, further investigation should be done. Once the current problems are dealt with (besides the pulley wobble, there's apparently a misfire issue), thread locking agent could certainly be used.

A 2.5's bolt can be tightened to the higher torque -- it's risky on a 2.2, which if I remember correctly, the one in question is. I'm wondering if there was some damage done when the pulley was originally loose.

Make sure there are no intake air leaks after the MAF. Any air that gets pulled in that isn't metered is air the ECU doesn't know about, and the ECU won't provide fuel in the right ratio. That can definitely affect idle quality.

There are two things I can think of offhand. The first is that air density changes with humidity level, so the situation might be related to an air metering issue. The second is that if the fuel you're using is slightly too low in octane rating, it might have a tendency to knock, and the ECU could retard the ignition timing a bit. When there's enough moisture in the intake air, it can somewhat reduce cylinder temperatures -- the cooling could reduce the tendency to knock, and allow the ECU to adjust timing to ''normal''.

''A little better'' wouldn't make me comfortable, either. If you think there's still too much wobble, reread the posts in this thread where I and others explained what the possible causes are. Yes, those are codes for misfire in the two cylinders at the front of the engine. They're of course in different banks, but one of the things they have in common is sharing an ignition coil in the pack. EDIT: If the CEL came on about the same time that the crank pulley wobble reappeared, it has me wondering if there's a problem in the area of the crank nose and key -- perhaps the crank sprocket isn't properly positioned, and the crank sensor isn't getting good pulses from the reluctor on the sprocket.

ECUs fail, but not too frequently. It could be very helpful to know what those ''random'' codes are. Try to get them and let us know.

My property is semi-wooded, with lots of wildlife and natural food sources here also. I keep my cars garaged, and had underhood damage occuring (some food storage and chewed wiring). I realized that the mice not only appreciated the warmth of a recently-run engine, but also liked the darkness and protection of the engine compartment. Initially I tried a Havahart trap baited with peanut butter, and had no problem catching mice. Since I didn't want to kill them, I released them at distance away that was supposedly well beyond that from which they should be able to find their way back. It didn't help -- either the same ones returned (don't know, since I didn't tag them ) or there were plenty to replace them. My later approach was twofold. First, I got a couple of the electronic repellers -- they're the kind that send out a periodic burst of an ultrasonic frequency, which we humans hear as a ''click''. Secondly, and probably more importantly, I prop the hood open as soon as I pull a car in. That tends to dissipate the heat more rapidly, so it's probably not as inviting to the critters, and takes away the dark/closed environment. I haven't had to use solder and heat-shrink tubing since.

The first thing to do is to try tightening the crank pulley bolt again. You do not need to loosen or remove any accessory drive belts to do that. If the wobble goes away again, it probably just indicates that you didn't tighten the bolt sufficiently last time.