OB99W

The sophisticated test gear helps, but isn't a necessity. Understanding how the EVAP system operates is key to fixing it. Coupling the under-dash TEST connectors will cause the solenoids to cycle with the engine running. Look at the label on the underside of the hood for what hoses and other parts are in the engine compartment. Using those two pieces of information, you should be able to at least verify the operation of the purge control solenoid by determining whether vacuum is being applied.

These EndWrench articles might be helpful: http://www.endwrench.com/current/02SpringInsiderInfo.pdf http://endwrench.com/pdf/fuel/MoreP0440TipsInfoW00.pdf http://www.endwrench.com/pdf/fuel/FtEvapEmissionTestW00.pdf

Is the idle different (speed, quality) if rather than using the service brake pedal, you apply the parking brake?

It's probably the ignitor or a bad connection to it. Otherwise, although unlikely, it could be due to an ECU failure. It should be possible to troubleshoot this with a voltmeter.

There are elongated holes in the sheetmetal below the plastic tray. The hooks go into those holes, and the rods will fall into the "grooves" of the tray. Perhaps you need to back off the nuts so that you can get the hooks to engage the holes. If you shift the tray back a bit (and have enough light) you should be able to easily see the front hole.

Crude oil is either paraffin-based, asphalt-based, or a combination of the two. Since you're partial to "dino oil", why do you prefer asphalt-based over paraffin-based? After 4 years, I wonder if they're still doing that.

May I suggest that you read this entire thread, including links to previous threads? You should be looking for a 6-position diagnostic connector (not all of the positions may have wires) that is often taped to the harness, along with two "diagnosis terminals" (ground wires that each have a pin that can mate with the diagnostic connector). In particular, see post #14 of this thread: http://www.ultimatesubaru.org/forum/showthread.php?t=96782&page=2

I suspect that it's going to be hard to get a definite answer to your question, unless someone else faced the same situation, tried it, and can tell you how things worked out. Having said that, since the last step is to turn the bolts through an angle (90 degrees) and not to a specific torque, it would seem that the delay shouldn't make too much difference in how tight they wind up being. If you decide to go ahead, be sure to do so in the right sequence. Of course, since I've never tried waiting a week between the last two steps of the procedure, don't blame me if it doesn't work out.

I just pasted those links, and for some reason they came out that color. Anyway, you're right, so I edited my post to fix it.

2.2L info in addition to other links: http://endwrench.com/images/pdfs/2.2SingleOverWin01.pdf http://endwrench.com/images/pdfs/2.2Liter.pdf http://endwrench.com/images/pdfs/TBeltEWWin05.pdf

The 16 flashes indicate an electrically-related problem -- see my post #8 here for some possibile causes: http://www.ultimatesubaru.org/forum/showthread.php?t=108790 If you get the code, let us know what it is and we can probably suggest a place to start looking.

Thanks for the info. The BR (black with red stripe) wire is pin #16 of the 16-pin B22. It's a ground connection for the ECU. B22 mates with E3. There are several other pins of that connector that carry ECU ground connections. Although EndWrench makes a point about this particular code and connector wire, in general poor connections between the "E" (engine) and "B" (body) harnesses, especially if they've been subjected to spray from salted roads, can be a cause of otherwise unexplained DTCs. Here's a link to the original EndWrench piece: http://endwrench.com/images/pdfs/SubaruLegacyInfo.pdf

Okay. For the P0135, heaters can fail in O2 sensors, but since the heater draws a lot of current, even a slightly poor contact can lead to a code. So, check connectors in that circuit. As to the P0441, the problem could be in the gas tank area, but if the code wasn't there before the engine work, start by checking hoses at the front, making sure things are connected as shown on the diagram stuck to the underside of the hood. Especially look at hoses related to the parts mounted on the passenger side strut tower. Trace the hose connection to the intake, and verify nothing was left disconnected.

You mentioned: P0135 (Front Oxygen Sensor Heater Circuit Malfunction) and P0441 (Evap System Incorrect Purge Flow) Before I get into possible causes, could you verify that second code? Because this one could indicate an electrical problem in common with P0135: P0141 (Rear Oxygen Sensor Heater Circuit Malfunction)

4S3 - World Manufacturer Identifier = USA, Subaru of Indiana, Passenger Vehicle B - Line = Legacy/Baja G - Body Style = Wagon 4 - Engine Type = 2.2L (for 1996-2000) 8 - Model = Outback/SUS 5 - Restraint = Manual Seat Belts, Driver & passenger Air Bags 3 - Check Digit (for verifying legitimacy of VIN) T - Model Year = 1996 6 - Assembly Plant & Transmission = Subaru Indiana, Full Time AWD 5MT 370419 - Sequence of Production I don't have info on determining the production date from those last six digits of the VIN, but that can be gotten from the label at the driver's door.

Either 8 valves per cylinder, or the engine's one of those rare Subaru 4.4L eights.

Separation of the two metal parts of the crank pulley due to failure of the rubber portion is not uncommon. Damage to the timing belt cover by rubbing of the pulley can happen from two causes -- either the pulley bolt being loose, or (as you found) the pulley coming apart. I'd suggest replacing it with an original part.

As long as you have all the correct sprocket marks at the correct positions and you're sure the tooth count is correct, then the belt will be properly installed and you can safely ignore the belt markings.

You're welcome. Congrats!

Yes, the crank pulley is located where you described, and you'd need a 22 mm socket. Normally the flywheel would be blocked to prevent the engine turning while tightening the bolt. At this point you might just verify that it's not loose without trying to hit full torque, since if the new belts eliminated the sound, then it's not likely that the crank pulley needs attention.

Assuming your parking brake is off... ...first make sure that you have a good, firm brake pedal, since bleeding doesn't always get all the air the first time. If that seems okay, sometimes a switch gets stuck -- try several rapid hard applications of the brake (while stationary, you don't have to be driving) and rapping lightly on the fluid reservoir. Let us know if the problem remains.

The bolt that retains the crank pulley is supposed to be tightened to about 130 ft-lbs. Let's try to pipoint where the noise is coming from. If you start the car when it's cold, and turn the steering wheel without moving the car, do you hear the sound? If so, get a spray bottle and put some water in it. With the engine off, spray the belt that drives the alternator and power steering pump (be careful - don't get water into the alternator). Start the engine and see if the sound is temporarily changed. Try the same with the belt that drives the A/C. Let us know what happens.

If you've had any engine work done recently that required the crankshaft pulley to be removed (timing belt, for example), make sure that the pulley isn't loose/wobbling. That can cause accessory belt squeal, and eventually a lot more trouble if not caught in time.

Your '98 should have a phase-1 4EAT, which didn't have the external spin-on filter that was used on the '99 and later phase-2 . The internal filter is coarse in comparison, so it doesn't tend to need much attention. I'd suggest just trying a few changes of fluid, especially since it's been a while since the last change, and see what happens.

Yes, if the AT Oil Temp light is blinking, that indicates an electrically-related problem that the TCU has detected, and it should be possible to extract one or more trouble codes. While a bad "C" duty solenoid can commonly cause torque bind (which in Subaru literature is usually referred to as "tight corner braking"), it isn't the only cause. Another possibility is a defective VSS2, although you'd typically expect the AT light to blink, and be able to get the code. Even problems with the throttle position sensor (TPS) can sometimes cause torque bind, since the ECU shares certain data with the TCU -- a "confused" TCU can misdirect things. Are there currently any drivability problems other than TB (which could suggest looking somewhere besides the usual culprits)?